Learning and memory function in young people with and without perinatal HIV in England.

Learning and memory are important for successful education and career progression. We assess these functions in young people (YP) with perinatal HIV (PHIV) (with or without a previous AIDS-defining illness) and a comparable group of HIV-negative YP. 234 PHIV and 68 HIV-negative YP completed 9 tests; 5 National Institutes of Health (NIH) Toolbox tests (2 executive function, 1 speed of information processing, 2 memory); 2 Hopkins Verbal Learning Test Revised (HVLT-R) (learning (L), delayed recall (R)), and 2 verbal application measures. Z-scores for each test were calculated using normative data and averaged by domain where appropriate. The effect of predictors on test scores in the three domains with the lowest z-scores were analysed using linear regression. 139(59%) and 48(71%) PHIV and HIV-negative YP were female, 202(86%) and 52(76%) Black, and median age was 19 [17, 21] and 18 [16, 21] years respectively. 55(24%) PHIV had a previous Center for Disease Control and Prevention (CDC) class C AIDS-defining diagnosis (PHIV/C). For HVLT-R, there was a trend towards PHIV/C YP having the lowest mean z-scores (L -1.5 (95% CI -1.8,-1.2), R -1.7 (-2.0,-1.4)) followed by PHIV without a CDC C diagnosis (L -1.3 (-1.4,-1.1), R -1.4 (-1.5,-1.2)) and then the HIV-negative group (L -1.0 (-1.3,-0.7), R -1.1 (-1.3,-0.8)); all were greater than 1 SD below the reference mean. The same trend was seen for verbal application measures; however, z-scores were within 1 SD below the reference mean. NIH Toolbox tests were similar for all groups. In multivariable analyses PHIV/C and Black ethnicity predicted lower HVLT-R scores. Black ethnicity also predicted lower executive function scores, however each year increase in age predicted higher scores. In conclusion, cognitive performance in verbal learning and recall fell below population normative scores, and was more pronounced in PHIV/C, supporting wider findings that earlier antiretroviral therapy initiation, before the occurrence of AIDS-defining conditions, may protect aspects of cognitive development.

Introduction

There are over two million children and adolescents living with HIV worldwide, the majority of whom acquired it perinatally [1]. Due to the availability of effective antiretroviral therapy (ART), young people living with perinatally acquired HIV (PHIV) are surviving to adulthood, but may have comorbidities resulting from lifelong HIV impacting brain development and neurocognitive functioning [2]. This is important, as cognitive abilities are key contributors to school performance and employability later in life. In the general child population, cognitive processes of executive function predict both reading comprehension and maths outcomes [3], and in PHIV young people, previous studies have shown that prospective and retrospective memory and executive function are key predictors of academic achievements [4].

Evidence from studies assessing neurocognitive status as well as results from neuroimaging in PHIV suggests that early HIV disease progression is associated with CNS impacts which endure into adolescence [5]. Many young people with PHIV who are alive today did not start HIV treatment in early life, under “treat all” approaches, and so potentially missed out on the benefits of this approach for cognitive outcomes [6]. A recent systematic review and meta-analysis found that working memory and executive function were the domains most affected by perinatal HIV infection, with processing speed also being affected [2].

We have previously reported similar performance in neuropsychological testing in young people living with HIV, without a prior history of AIDS-defining complications, and a comparable group of HIV negative (HIV-) young people (YP) within the Adolescents and Adults Living with Perinatal HIV (AALPHI) cohort study [7]. Of note, PHIV with a past history of AIDS-defining conditions showed poorer performance in all cognitive function domains explored. In this paper we report on further cognitive testing performance of young people with PHIV (with or without a previous AIDS-defining illness) and a comparable group of HIV- young people in function domains considered particularly relevant to school progression and transition to adulthood. We explored six function domains with particular focus on executive function, verbal learning and memory because these cognitive functions have been associated with educational and employment outcomes. We also examined predictors of cognitive performance in the three domains with the lowest scores, and additional analyses restricted to participants with PHIV were conducted to investigate the effect of HIV-related factors. For this analysis, we included neuropsychological tests widely used in similar population groups with the aim of generating results easily interpretable in the context of previous studies.

Methods

The AALPHI cohort was a prospective study evaluating the impact of HIV infection and ART exposure on young people with PHIV and comparing them to HIV- young people. PHIV acquired the infection perinatally whereas HIV- YP were affected by HIV (mainly siblings of PHIV participants or a young person with a parent living with HIV), thus providing a comparison group which had similar background environmental and psychosocial factors related to HIV exposure. Participants were enrolled from HIV clinics and community service locations in England and undertook first interviews between 2013 and 2015, and second interviews between 2015 and 2017. Detailed methods have been described previously [7]. All participants with PHIV had known their HIV status for at least six months, were aged 12–21 years, and were all included in the national UK and Ireland Collaborative HIV Paediatric Study (CHIPS) [8]. HIV- young people were aged 13–23 years and had either a sibling or parent living with HIV. Young people who had been living in the UK for less than six months or were unable to speak or understand English were excluded from the study. Participants provided written informed consent, except where they lacked the capacity, in which case parents or guardians provided written informed consent and the young person provided written assent. Young people younger than 18 years were allowed to consent to participation in the study themselves if they were deemed by the study research nurses as having the capacity to consent. Full ethical approval was obtained from Leicester Research Ethics Committee.

At the participant’s second two-hour face to face interview by research nurses, cognitive assessments (consisting of nine tests exploring six function domains) were carried out. Training was provided on all the cognitive assessments by the study psychologist. Five tests were administered via the computerised National Institutes of Health (NIH) Toolbox® cognition battery (Executive function: flanker inhibitory control and attention test and dimensional change card sort test, Speed of Information Processing: pattern comparison test, and Memory: list sorting and picture sequence test) [9]. Four tests were paper-based, including 2 Hopkins Verbal Learning Test-Revised™ (HVLT-R™) tests for verbal learning (sum of three trials) and verbal delayed recall [10], and 2 verbal application tests: Weschler Individual Achievement Test® Second UK Edition (WIAT®-IIUK) (a word reading task) [11] and British Picture Vocabulary Scale Third Edition (BPVS III) (a word comprehension task) [12].

Data were analysed using STATA version 15.1 (Stata Corp, College Station, Texas, USA). Scoring of all tests followed manufacturer recommendations. For each test, z-scores were calculated using normative data (NIH Toolbox® and HVLT-R™ normative data were adjusted for age, sex, ethnicity and years of education [13, 14], and WIAT®-IIUK and BPVS normative data adjusted for age). Tests designed and validated to assess cognitive function in a domain were grouped and Cronbach α was used as a measure of internal consistency (the lower bound set at 0.6). For domains with more than 1 test, z-scores were averaged to give mean z-scores for each domain. For each domain, mean z-scores were compared across HIV-, PHIV without a CDC clinical classification class C past history of an AIDS-defining illness [15, 16] (PHIV/no C) and PHIV with a CDC C diagnosis (PHIV/C) groups using ANOVA tests. The normality of the mean z-scores was assessed using visual plots and Sharpio-Wilk tests. Characteristics of participants by HIV and CDC status were compared using Chi squared or Fisher’s exact tests for proportions (as appropriate given the number of cells with an expected value of five or greater), and Wilcoxon rank sum (comparing two groups) or Kruskal-Wallis (comparing three groups) tests for medians. Results are presented for non-missing values; missing values were less than 10% of study participants unless specified. Ethnicity was self-reported by study participants in the following categories (used by the UK government): White (British, Irish, other), Asian or Asian British (Indian, Pakistani, Bangladeshi, Chinese, other), Black or Black British (Caribbean, African, other), Mixed (White and Black Caribbean, White and Asian, White and Black African, other), Other, Prefer not to say. Ethnicity was analysed as Black (Black or Black British) or non-Black (all other categories) due to small numbers.

The effect of potential predictors on test scores in the three domains with the lowest mean z-scores overall were analysed using linear regression. Factors considered a priori by the AAPLPHI Steering Committee to be associated with test scores were HIV status and CDC disease stage (HIV-, PHIV/no C, PHIV/C), sex, age, ethnicity and country of birth (within vs. outside the UK/Ireland). Other variables considered to be important as potential predictors are shown in Table 1 (denoted with a * asterisk). An additional analysis for only participants with PHIV was also conducted to investigate the effect of HIV-related factors on cognitive performance (HIV-related indicators considered as predictors are shown in Table 2 with a * asterisk; pre-AALPHI clinical data were collected through the Collaborative HIV Paediatric Study). Variables with a P-value <0.15 in univariable analyses were considered in multivariable analyses using backwards selection, as well as the a priori variables, and a P-value <0.05 was considered statistically significant.

Table 1

Characteristics of participants by HIV and CDC status.

	HIV-	PHIV			P-value
	HIV- (n = 68)	No CDC C (n = 179)	CDC C (n = 55)	Total PHIV (n = 234)	HIV- vs PHIV	HIV- vs PHIV/no C vs PHIV/C	PHIV/no C vs PHIV/C
Sociodemographics:
Male sex, No. (%)	20 (29%)	68 (38%)	27 (49%)	95 (41%)	0.094	0.082	0.143
Age at interview 2, No. (%)
≤15 years	14 (21%)	22 (12%)	9 (16%)	31 (13%)
16–18 years	26 (38%)	76 (43%)	23 (42%)	99 (42%)	0.325	0.586	0.725
≥19 years	28 (41%)	81 (45%)	23 (42%)	104 (45%)
Median [IQR]	18 [16,21]	19 [17,21]	18 [17, 21]	19 [17, 21]	0.453	0.600	0.474
Years between interviews, median [IQR]	1.7 [1.3,2.4]	2.0 [1.4,2.6]	1.8 [1.4, 2.7]	2.0 [1.4, 2.6]	0.082	0.220	0.945
Ethnicity (Black), No. (%)	52 (76%)	154 (86%)	48 (87%)	202 (86%)	0.050	0.144	0.815
Born outside UK/Ireland, No. (%)	30 (44%)	112 (63%)	24 (44%)	136 (58%)	0.040	0.006	0.013
Psychosocial:
Death of one/both parents * , No. (%)	16 (26%)	73 (44%)	19 (36%)	92 (42%)	0.021	0.039	0.297
Living situation * , No. (%)
Family own or rent property	23 (34%)	66 (37%)	22 (40%)	88 (38%)
Housing association/council property	36 (53%)	86 (48%)	29 (53%)	115 (49%)	0.834	0.669	0.365
Other	9 (13%)	26 (15%)	4 (7%)	30 (13%)
Occupation * , No. (%)
Education	51 (75%)	132 (74%)	43 (78%)	175 (75%)
Employment	11 (16%)	34 (19%)	7 (13%)	41 (18%)	0.895	0.808	0.502
Not in education, training or employment	6 (9%)	12 (7%)	5 (9%)	17 (7%)
Years of education, median [IQR]	11 [10,14]	11 [10, 14]	11 [10, 14]	11 [10, 14]	0.502	0.742	0.664
Highest level of parent/carer education* a, No. (%)
Did not complete high school	4 (7%)	8 (5%)	4 (9%)	12 (6%)
High school graduate	25 (44%)	59 (41%)	20 (47%)	79 (42%)	0.933	0.770	0.429
University graduate	28 (49%)	79 (54%)	19 (44%)	98 (52%)
Ever excluded from school * , No. (%)	17 (25%)	37 (21%)	13 (24%)	50 (22%)	0.583	0.804	0.711
Environmental:
Ever fostered/adopted* b, No. (%)	1 (2%)	14 (13%)	4 (12%)	18 (13%)	0.049	0.118	1.000
Number of adult carers* c (excluding parents)
Median [IQR]	0 [0, 1]	0 [0, 1]	0 [0, 1]	0 [0, 1]	0.862	0.472	0.232
Main language spoken at home* d, No. (%)
English only	30 (44%)	92 (52%)	27 (49%)	119 (51%)
English and another equally	34 (50%)	82 (46%)	25 (45%)	107 (46%)	0.388	0.535	0.473
Languages other than English	4 (6%)	4 (2%)	3 (5%)	7 (3%)
Young carer * , No. (%)	21 (31%)	16 (9%)	7 (13%)	23 (10%)	<0.001	<0.001	0.424
Lifestyle:
Ever smoked * , No. (%)	27 (40%)	72 (41%)	24 (44%)	96 (42%)	0.854	0.871	0.623
Ever consumed alcohol * , No. (%)	44 (65%)	117 (65%)	27 (50%)	144 (62%)	0.664	0.113	0.042
Waist to hip ratio * , median [IQR]	0.9 [0.8,0.9]	0.9 [0.8,0.9]	0.9 [0.8, 0.9]	0.9 [0.8, 0.9]	0.397	0.203	0.122
Waist to hip ratio z-score e, median [IQR]	1.2 [0.2, 2.1]	1.2 [0.3,2.0]	1.4 [0.6,2.4]	1.3[0.4,2.1]	0.665	0.544	0.300
Mental health:
EQ-5D-5L Anxiety/Depression* f, No. (%)
No anxiety or depression	41 (61%)	109 (62%)	30 (57%)	139 (61%)
Slightly/moderately anxious/depressed	21 (31%)	56 (32%)	18 (34%)	74 (32%)	0.983	0.930	0.657
Severely/extremely anxious/depressed	5 (8%)	11 (6%)	5 (9%)	16 (7%)

Abbreviations: HIV-, HIV negative; IQR, interquartile range; PHIV, perinatal HIV; PHIV/C, PHIV with a CDC C diagnosis; PHIV/no C, PHIV without a CDC C diagnosis

* variables considered as predictors in multivariable models (as well as HIV status and CDC disease stage, sex, age, ethnicity and country of birth as a priori variables)

a unknown for 11 HIV-, 33 PHIV/no C, 12 PHIV/C;

b unknown for 25 HIV-, 73 PHIV/no C, 21 PHIV/C (question was introduced part way through interview 2);

c number of adults taking responsibility for and living with the young person during childhood excluding parents;

d at the time of interview 1;

e calculated with reference to healthy uninfected children, adjusted for age and sex [17];

f anxiety/depression dimension of the EQ-5D-5L instrument was used to measure anxiety and depression [18]

Table 2

HIV-related characteristics of PHIV participants by CDC status.

	PHIV			P-value
	No CDC C (n = 179)	CDC C (n = 55)	Total (n = 234)	PHIV/no C vs PHIV/C
Age at first presentation * , No. (%)
Birth	14 (8%)	4 (7%)	18 (8%)
<1 year	16 (9%) 52 (25%)	16 (30%)	32 (14%)
1–4 years	49 (27%)	14 (26%)	63 (27%)	0.003
5–9 years	58 (32%)	13 (24%)	71 (30%)
≥10 years	42 (23%)	7 (13%)	49 (21%)
Median [IQR]	6.0 [2.1,9.4]	2.2 [0.3,7.1]	5.2 [1.5,9.2]	0.005
Year of first presentation * , No. (%)
Up to 1996	28 (16%)	9 (17%)	37 (16%)
1997–2000	32 (18%)	20 (37%)	52 (22%)	0.008
2001 onwards	119 (66%)	25 (46%)	144 (62%)
ART status at interview * , No. (%)
Never started ART	12 (7%)	0 (0%)	12 (5%)
On ART	155 (86%)	52 (95%)	207 (89%)	0.135
Off ART (previous ART exposure)	12 (7%)	3 (5%)	15 (6%)
Age at initiation of ART * , median[IQR] (on or off ART)	9.2 [4.8,12.8]	3.6 [0.5,8.5]	7.5 [3.4,12.2]	<0.001
Year initiated ART * , No. (%) (on or off ART)
Up to 1996	11 (7%)	4 (7%)	15 (7%)
1997–2000	25 (15%)	25 (46%)	50 (22%)	<0.001
2001 onwards	131 (78%)	26 (47%)	157 (71%)
No. of drugs at ART initiation * , No. (%) (on or off ART)
1 or 2	19 (11%)	7 (13%)	26 (12%)	0.787
3 or 4	148 (89%)	48 (87%)	196 (88%)
Class at ART initiation * , No. (%) (on or off ART)
NNRTI based	107 (64%)	31 (56%)	138 (62%)
Boosted PI based	28 (17%)	7 (13%)	35 (16%)	0.182
Unboosted PI based/NRTI only	32 (19%)	17 (31%)	49 (22%)
Taking efavirenz at interview * , No. (%) (on ART only)	37 (24%)	9 (17%)	46 (22%)	0.325
Viral load <50 copies/ml at interview* a, No. (%) (on ART only)	101 (71%)	32 (64%)	133 (71%)	0.221
Median[IQR] cumulative years viral load <400 copies/ml *	6.9 [3.9,10.0]	6.9 [4.0,11.9]	6.9 [3.9,10.2]	0.348
Median[IQR] cumulative years viral load <50 copies/ml * <50 copies/ml**	6.0 [2.7,9.1]	6.2 [2.9,10.4]	6.1 [2.8,9.3]	0.524
CD4 nadir (cells/mm 3 ) * , medium [IQR]	234 [168,380]	166 [17,330]	234 [131,359]	0.003
CD4 at interview* a,b (cells/mm3), medium [IQR]	555 [402,784]	641 [485,776]	581 [407,782]	0.260

Abbreviations: IQR, interquartile range; PHIV, perinatal HIV; PHIV/C, PHIV with a CDC C diagnosis; PHIV/no C, PHIV without a CDC C diagnosis; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; NRTI, nucleoside reverse transcriptase inhibitor.

* additional variables considered as predictors in multivariable models in the PHIV group only.

a within 6 months before or after interview;

b unknown for 33 PHIV/no C, 6 PHIV/C

Results

A total of 234 PHIV and 68 HIV- young people completed cognitive testing at interview 2 and 55 (24%) of PHIV young people had a CDC C diagnosis. There were more females than males in each group, most were Black and around half were born outside of the UK/Ireland (Table 1). The median age was 19 (IQR 17,21) and 18 (16,21) years (p = 0.453) for the PHIV and HIV- groups respectively and the majority of young people were still in education in each group. For 42% of PHIV and 26% of HIV- YP one or both parents had died, more PHIV young people had been fostered or adopted (13% PHIV, 2% HIV-) and more HIV- young people were young carers (10% PHIV, 31% HIV-). There were no other statistically significant differences between the groups in psychosocial, environmental or mental health characteristics (Table 1). In the PHIV group, those with a CDC C diagnosis were more likely to have presented to HIV care and initiated ART at a younger age and in earlier calendar years, than those without a CDC C diagnosis (Table 2). Most had started ART with 3 or 4 drugs and a regimen containing a non-nucleoside reverse transcriptase inhibitor (NNRTI) or boosted protease inhibitor (PI).

Cognitive function mean test and z-scores by domain, HIV and CDC status are shown in S1 Table. Mean z-scores and 95% confidence intervals were below reference means in nearly all HIV/CDC groups and domains, suggesting lower cognitive performance compared with normative data (Fig 1). For verbal learning (L) and verbal delayed recall (R), there was a trend towards PHIV/C young people having the lowest mean z-scores (L -1.5 (95% CI -1.8,-1.2), R -1.7 (-2.0,-1.4)) followed by PHIV/no C (L -1.3 (-1.4,-1.1), R -1.4 (-1.5,-1.2)) and then HIV- young people (L -1.0 (-1.3,-0.7), R -1.1 (-1.3,-0.8)) (L p = 0.026, R p = 0.004), and all were greater than minus one standard deviation below the reference mean. A similar trend was seen for verbal application measures (PHIV/C -0.5 (-0.7,-0.3), PHIV/no C -0.2 (-0.4,-0.1), HIV- -0.1 (-0.3,0.1), p = 0.034), however they were within one standard deviation below the reference mean for all groups, suggesting that these skills were less compromised than verbal learning and verbal delayed recall. Executive function, speed of information processing and memory tests z-scores were similar for all three groups.

Fig 1

Cognitive performance by domain and HIV/CDC status.

Data points are means with 95% confidence intervals. P-values compare the three groups (HIV, PHIV/no C, PHIV/C) using analysis of variance. Abbreviations: CDC, Centers for Disease Control and Prevention, HIV-, HIV negative; PHIV, perinatal HIV; PHIV/C, PHIV with a CDC C diagnosis; PHIV/no C, PHIV without a CDC C diagnosis.

Multivariable analyses looked at predictors of improved verbal learning (L), verbal delayed recall (R), and executive function (Flanker inhibitory control and attention (F), and dimensional change card sort (D)) scores (S2 Table). Both before and after adjustment for other factors, verbal learning and verbal delayed recall scores were lower in PHIV/C young people (multivariable coefficient L -2.55 (95% confidence interval -4.25, -0.85); R -1.18 (-2.10, -0.25), p = 0.040) and young people of Black ethnicity (R -2.59 (-4.11, -1.06); L -1.53 (-2.34, -0.73), p<0.001). Executive function scores were also lower in Black ethnicity young people (F -0.30 (-0.58, -0.01), p = 0.040; D -0.35 (-0.67, -0.04), p = 0.029) and improved with each year increase in age (F 0.04 (0.00, 0.08), p = 0.033; D 0.09 (0.04, 0.13), p<0.001). There were also some factors which were only predictors for one of the four tests in multivariable analysis; higher waist to hip ratio was associated with lower verbal delayed recall scores (-4.76 (-8.63, -0.88) per 0.1 increase in waist to hip ratio, p = 0.016), ever having smoked was associated with higher flanker inhibitory control and attention scores (0.22 (0.01, 0.43), p = 0.039), and loss of a parent was associated with lower dimensional change card sort scores (-0.32 (-0.56, -0.09), p = 0.007) as well as being a young carer (-0.39 (-0.72, -0.05), p = 0.024).

A separate model for the PHIV group found similar results to the overall model in terms of associations with CDC status, Black ethnicity and age (S3 Table). In multivariate analysis, having a supressed viral load <50 copies/ml at the time of the interview was associated with higher verbal learning scores (multivariable coefficient 2.25 (0.90, 3.61) p = 0.001) and starting ART with a regimen containing a boosted PI was associated with higher verbal delayed recall scores (1.48 (0.51, 2.45) p = 0.011).

Discussion

In this well characterised cohort of young people with perinatally acquired HIV infection and a comparable group of HIV- young people, we found similar levels of performance across the three HIV/CDC groups and for most of the cognitive domains. However, performance was lowest in the verbal learning and delayed recall domains, and the lowest performance was consistently seen in PHIV/C young people. The scores of PHIV/C young people on verbal application measures were also lower than in the other two groups, but the mean z-score for all groups were within 1 SD below the normative mean.

Poorer cognitive performance in PHIV/C young people has been observed in other studies. Malee et al found PHIV with past AIDS-defining diagnoses had lower performance across multiple executive function tasks than PHIV/no C and perinatally HIV-exposed uninfected youth (PHEU) [19]. In another study, 45 PHIV/C had significantly lower visual recognition memory compared with 85 PHEU [20]. A study in south-east Asia reported PHIV performing significantly worse on visual memory tests when compared with PHEU at baseline, but found no difference between groups in the trajectory of working memory measures over the three year follow-up period [21].

Consistent with previous reports, performance of the entire cohort in most cognitive domains was lower than normative population means and may be lower than we have reported, as those with severe cognitive impairment may not have been referred to the study. This suggests that performance in cognitive tasks may be tempered by factors and influences not related to HIV disease and highlights the importance of having an appropriate comparison group [4, 19, 22]. However, a longitudinal analysis comparing cognitive testing performance in children with perinatally acquired HIV and HIV-exposed but uninfected children in Zambia, recently reported that socioeconomic factors impacted testing performance in both groups, but with a more marked effect in the PHIV group [23]. Another study in South Africa showed uninfected children exposed to maternal HIV infection had language delays at 2 years of age, so further work is needed to understand developmental outcomes of this population [24].

Executive function scores improved with increasing age but age was not associated with improved verbal learning or delayed recall scores. The relative contribution of specific cognitive abilities may vary according to age, and previous studies have shown that working memory and concrete and visual reasoning dominate in primary school whereas language and more abstract reasoning appear to do so in secondary school [25, 26]. Abdominal obesity was associated with poorer verbal delayed recall, but not other cognitive function domains. Abdominal obesity has been associated with poorer cognitive performance in adults living with HIV [27] and in children in the general population [28, 29]. However, to our knowledge no data have been published on this association in young people living with HIV and further research is needed.

In contrast with findings at the first interview on this cohort, which whilst used different cognitive assessments measured similar domains [7] (tests at the second interview were easier to administrate and had more appropriate normative data for our population), two years later we observed no difference in performance between PHIV with and without CDC C diagnoses in speed of information processing, memory and executive functioning. This may well reflect ongoing brain development during adolescence [30] in the absence of significant further HIV-related insults such as opportunistic infections, and is consistent with previous studies in which cognition seems to improve [21, 29] or to remain stable over time in the short term [19]. We detected an effect of ethnicity on cognition in this analysis, as we did in our previous paper. Although the NIH Toolbox and HVLT-R normative data are adjusted for ethnicity, their normative sample was around 60% and 58% White, and 15% and 42% African American respectively, and this may contribute to the difference we found [13, 14].

Our study has a number of limitations. Firstly, the tools we used to investigate cognitive performance differed at the different study visits and therefore we only have cross-sectional data on the associations between these measures and HIV status. However, if practice effect was a factor to be considered, using different tests to assess performance on a given cognitive domain may be advantageous. Secondly, the study was unable to ascertain any effect of different antiretroviral drug or drug class on cognition.

In conclusion, our study showed similar performance in cognitive testing between PHIV and HIV- young people in most function domains. However, performance in verbal learning and recall fell below population normative scores, and was more pronounced in PHIV/C, supporting wider findings that early initiation of ART, before the occurrence of AIDS-defining conditions, may protect aspects of cognitive development.

Cognitive function test and z-scores by domain, HIV and CDC status.

Mean scores were compared across the three groups (HIV-, PHIV/no C, PHIV/C) using ANOVA tests. Abbreviations: HIV-, HIV negative; PHIV, perinatal HIV; PHIV/C, PHIV with a CDC C diagnosis; PHIV/no C, PHIV without a CDC C diagnosis.

(DOCX)

Click here for additional data file.

Predictors of improved verbal learning, verbal delayed recall and executive function scores.

All a priori variables, as well as those with univariable P-value <0.15 and multivariable P-value <0.05 are presented. Abbreviations: CI, confidence interval, HIV-, HIV negative; PHIV, perinatal HIV; PHIV/C, PHIV with a CDC C diagnosis; PHIV/no C, PHIV without a CDC C diagnosis.

(DOCX)

Click here for additional data file.

Predictors of improved verbal learning, verbal delayed recall and executive function scores for PHIV participants.

All a priori variables, as well as those with univariable P-value <0.15 and multivariable P-value <0.05 are presented. Abbreviations: CI, confidence interval, HIV-, HIV negative; PHIV, perinatal HIV; PHIV/C, PHIV with a CDC C diagnosis; PHIV/no C, PHIV without a CDC C diagnosis; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; NRTI, nucleoside reverse transcriptase inhibitor. a within 6 months before or after interview.

(DOCX)

Click here for additional data file.

6 Dec 2021

Learning and memory function in young people with and without perinatal HIV in England

PLOS ONE

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Lori Buchanan, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

3. One of the noted authors is a group or consortium "Adolescents and Adults Living with Perinatal HIV (AALPHI) Steering Committee". In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

4. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for the opportunity to review this manuscript. The authors sought to compare the cognitive functioning of three groups: people with PHIV and CDC C diagnosis, people with PHIV and no CDC C diagnosis, and HIV- individuals with a sibling/parent who has HIV.

This is a clinically relevant and important issue for a wide variety of clinicians. Understanding cognitive functioning in these groups has important implications for clinical service allocation, school/work accommodations, and other support. The HIV- control group is a strength of this study. However, the manuscript would benefit from a more thorough review of the literature; clarification of methods; and a more nuanced discussion of cognitive test findings. Please find my suggestions below.

Major Revisions

1. The manuscript requires better alignment of aims, methods, and results. For example, the aim in the abstract states, “we assess these functions in young people (YP) with perinatal HIV (PHIV) and a comparable group of HIV-negative (HIV-) YP.” However, the methods described three groups. Additionally, the aim is to compare groups on cognitive functioning, but a regression model is run to assess predictors of cognitive functioning. No objective or hypothesis for the regression analysis is reported, nor are the results of the regression reported in the abstract.

2. Overall, the findings need to be better contextualized within the broader literature in both the introduction and the discussion. For the introduction, what is already known about the effects of perinatal HIV on cognition in young people? What don’t we know about cognitive functioning in this population? Why is this study important? Moreover, the authors cite a previously published paper from their own working group on cognitive performance in the same cohort. These findings should be reported in the introduction. It is unclear how this study differs from their previously published paper.

3. For domains with more than one test, z scores were averaged to create a mean z score for each domain. This seems appropriate; however, the rationale for combining certain tests needs to be made explicit. From a neuropsychological perspective, some of the combinations may not be appropriate. For example:

a. What is the rationale for combining list sorting and picture sequence from the NIHTB? The list sorting test is a verbal working memory task and picture sequence is a visual episodic memory task. Conceptually, a working memory task might be better defined as an executive functioning task. See https://pubmed.ncbi.nlm.nih.gov/24959983/.

b. The picture memory task shows good convergent validity with the RAVLT (a very common list learning test), which is similar to the HVLT (see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4254833/). What is the rational for calling the HVLT a learning task and separating it from the other memory tasks? Learning is a component of memory. If the HVLT is impaired this is a component of verbal memory.

4. Although the results for the Verbal Learning PHIV C vs PHIV-C are statistically different, the confidence intervals are completely overlapping. Moreover, the z-scores are very similar (1.3 vs 1.5). This translates to a standard score of 78 vs 81, respectively. Is this really clinically meaningful? These scores may result in very similar behavioral presentation. Effect sizes should be reported to demonstrate magnitude of effects.

5. More clarity is warranted with regard to the aim and methods of the regression analysis:

a. What is the goal of the regression analysis? If part of the primary objective of the paper is to determine what factors might predict poorer cognitive functioning in this group, that should be made explicit in the introduction/aims.

b. IQ/education are important predictors of cognitive functioning. Were these controlled for in the regression to determine what factors predict poor outcomes? If not, why?

c. The authors note that the predictors used in the regression can be found in Table 1. Table 1 has over 20 variables. Are all of these included in the regression? I would think the analysis would be underpowered for that many variables. Please clarify and provide justification for the predictors.

d. An additional analysis with only PHIV was conducted. What is the rationale for this? Was this decided a priori or is this a sensitivity analysis? This needed clarification.

6. Raw score means, SDs, and ranges for all cognitive testing should be presented. This is necessary for interpretation of the results.

Minor Revisions by Section

Abstract

1. The title and Abstract background make it sound like only learning and memory was assessed. Consider updating both.

2. Acronyms are not defined before use. (e.g., CDC C diagnosis, ART)

3. The original aim is to compare PHIV to HIV-PV, but there is a three-group comparison reported in results.

4. Results of the linear regressions are not mentioned.

5. The conclusion made in the abstract that results suggest earlier ART leads to better outcome is not support by the results reported in the abstract. There is no mention of ART in the result section.

Introduction

6. The connection to education in the introduction is a little unclear. This might be an important implication of the work; however, it is not directly related to the study aim.

7. A rationale should be provided for using the HIV-negative control group. I assume this control group accounts for environmental and psychosocial factors related to HIV exposure. This should be made explicit.

8. In line with the above, there is clearly good reason for comparing groups with and without a CDC C diagnosis separately, but the rationale should be made explicit.

9. Why are no a prioiri hypotheses reported?

Method

10. The first sentence needs editing. It reads like the study was evaluating the impact of HIV infection and ART on HIV- young people as it is currently written.

11. Please report any exclusion criteria for the larger study.

12. A lot of space is used to discuss the consent/assent process. This is good/ethical practice, but unnecessarily detailed for a manuscript.

13. Cognitive testing described to have occurred during a face-to-face interview. It may be better described as an interview with cognitive testing.

14. Please define the CDC Clinical classification class C past history of an AIDS defining illness.

Statistical analysis

15. Specify what comparisons are being made using chi square tests/Wilcoxon/KW.

Results

16. Were there any significant differences between your groups on important demographic or clinical factors such as education, SES, psychiatric diagnoses? I believe these comparisons are made in Table 1, but the authors should include a statement in the results and then point readers to the table.

17. Information on variable distribution/assumptions underlying statistical analyses should be reported.

Discussion

18. It would be interesting to discuss in more detail why your control group also performed lower than the normative sample on several tests to contextualize the results. For example, what common factors between all three groups could have impacted cognitive performance on these measures and is this consistent with the literature?

19. Based on what is reported in the discussion, the HIV+ cohort in this study had impairments in cognitive domains that differed from previously published worked. This should be discussed.

20. There is a note to add “blackmore reference” in the discussion.

21. It is unclear why the second limitation is a limitation of the study. Was identifying differential effect of antiretroviral drugs on cognition a goal? Is there a reason to believe that different treatments affect cognition in different ways? If so, this should be made more explicit.

22. A limitation of this study is only using active recall to test memory (i.e., no recognition memory test from HVLT), therefore it is unclear if deficits in delayed memory extend to recognition. If this cohort has impaired learning and recall, but not encoding (i.e., recognition memory is average), this would suggest a frontal executive, as opposed to amnestic mesial temporal profile.

23. The authors conclude that their findings support that earlier ART protects against cognitive development. Was age of ART included in the regression model? Did it predict outcomes? It seems you have the data to test this assumption, but it is not clearly reported anywhere in the manuscript.

Tables

24. Table 1 is very large. I would recommend splitting up into more than one table.

25. Should present results of the linear regressions in a table for transparency.

Reviewer #2: Abstract

1. Background paragraph of abstract emphasizes relevance of learning and memory for successful educational and career progression, but this was not examined in this study. Authors should consider a more relevant background justification that would more directly apply to this work and supporting analyses.

2. “CDC C” diagnosis term in abstract would not be known to a general scientific audience. Consider briefly defining.

Methods

1. The section on the consent process can be shortened/streamlined.

2. It would be helpful for the authors to briefly explain why it would be relevant to compare PHIV/no C and PHIV/C. This is done within the discussion, but it would be beneficial for the reader to understand the reasoning for this comparison from the outset (perhaps within the introduction?)

3. Executive functioning and memory performance can be directly impacted by attention, and attention difficulties are documented in HIV. Was there a reason attention measures were not included in this battery?

Results

1. While it is logical that multiple test scores within a domain would be averaged to present results in a cohesive manner, the authors should make explicit that the scores were sufficiently statistically similar for this to be appropriate. Please include a table summarizing mean performance per group for all individual tests using a common metric (e.g., z scores). Also explain the criteria for averaging across tests within a domain; for example, did the authors ensure that mean test scores within a domain were not statistically different from each other? I imagine that there may have been some variability in performance for within-domain tests. Is it possible that potential effects were attenuated because of variable performance within domains?

Discussion

1. There were some lower cognitive scores amongst participants of Black race found in adjusted multivariable models. It is important that the authors address potential explanations for this within the discussion.

General comment for consideration:

1. The term “Black” generally refers to a race/racial category, as you can have participants who are “Black” but report different ethnic backgrounds. The authors can consider these recently published suggestions for reporting of race or ethnicity: https://jamanetwork.com/journals/jama/fullarticle/2776936

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

4 Mar 2022

PONE-D-21-16526

Learning and memory function in young people with and without perinatal HIV in England

PLOS ONE

We thank the journal and reviewers for their time considering our paper. We are delighted with the opportunity of submitting a revised version of our manuscript for your consideration and, following your request, we have addressed the comments made by your editorial office and the reviewers below.

Academic editor comments:

1. The analyses section needs reconsideration and cleaning up. All data should be provided so that readers can make their own determination and the combining of test scores should be given more thought. This was seen as a minor revision by Reviewer one and as a major revision by Reviewer 2 but both believed that this aspect of the study could be much improved and that a more sophisticated and nuanced approach to test interpretation was needed.

Response: More detail has been added to the methods and results sections, including the addition of supplementary tables detailing the raw data from the individual tests and results of the regression analyses. We have also added more details on the rationale of combining some of the tests into domains.

2. Both reviewers also questioned whether the introduction adequately positioned the study within the literature and reviewer 2 wanted an indication of what this study adds to your previous reporting of the same data set.

Response: We have added more details to the introduction about our previous study looking at cognitive function at interview 1.

Please see our responses to individual comments below.

Journal Requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Response: This has been done.

2. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

Response: The corresponding author is affiliated to two Institutes from University College London, the Institute of Clinical Trials and Methodology (ICTM) and the Institute for Global Health (IGH). The MRC Clinical Trials at UCL is based at ICTM. It is not clear to us where the confusion is, but would be happy to work with you to get this sorted.

3. One of the noted authors is a group or consortium "Adolescents and Adults Living with Perinatal HIV (AALPHI) Steering Committee". In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

Response: The membership and lead contact for the AALPHI steering committee is already stated in the Acknowledgements section. We have added a note to the title page to indicate that membership of the group is provided in the Acknowledgements.

4. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

Response: The data have now been included in a Supporting Information file (S3 Table).

Reviewers' comments:

Reviewer #1:

Major Revisions

1. The manuscript requires better alignment of aims, methods, and results. For example, the aim in the abstract states, “we assess these functions in young people (YP) with perinatal HIV (PHIV) and a comparable group of HIV-negative (HIV-) YP.” However, the methods described three groups. Additionally, the aim is to compare groups on cognitive functioning, but a regression model is run to assess predictors of cognitive functioning. No objective or hypothesis for the regression analysis is reported, nor are the results of the regression reported in the abstract.

Response: We have amended the abstract and introduction to better reflect the three groups compared in the analysis. We have also included the regression analysis in the abstract and added it as an aim in the introduction.

2. Overall, the findings need to be better contextualized within the broader literature in both the introduction and the discussion. For the introduction, what is already known about the effects of perinatal HIV on cognition in young people? What don’t we know about cognitive functioning in this population? Why is this study important? Moreover, the authors cite a previously published paper from their own working group on cognitive performance in the same cohort. These findings should be reported in the introduction. It is unclear how this study differs from their previously published paper.

Response: As the reviewer mentioned, we have explored cognitive function performance in the AALPHI cohort at baseline. In the initial assessment we found no overall difference in neuropsychological testing performance between young people living with HIV, who have never developed any AIDS-defining complication, and HIV-negative young people. However, performance on tests exploring memory and information processing cognitive domains seem to be poorer in PHIV, particularly those who have experienced AIDS-defining complications in the past. Therefore, the round of neuropsychological assessment we are presenting in this manuscript, focused predominately on learning and memory cognitive function domains because of their implications on educational and employment outcomes.

We have amended the introduction to reflect this.

3. For domains with more than one test, z scores were averaged to create a mean z score for each domain. This seems appropriate; however, the rationale for combining certain tests needs to be made explicit. From a neuropsychological perspective, some of the combinations may not be appropriate. For example:

a. What is the rationale for combining list sorting and picture sequence from the NIHTB? The list sorting test is a verbal working memory task and picture sequence is a visual episodic memory task. Conceptually, a working memory task might be better defined as an executive functioning task. See https://pubmed.ncbi.nlm.nih.gov/24959983/.

b. The picture memory task shows good convergent validity with the RAVLT (a very common list learning test), which is similar to the HVLT (see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4254833/). What is the rational for calling the HVLT a learning task and separating it from the other memory tasks? Learning is a component of memory. If the HVLT is impaired this is a component of verbal memory.

Response: We do agree that the findings do not clarify the specific neurocognitive processes most affected by the presence of HIV in the developing brain. The choice of tests recognised the complex interaction between attention, memory, information processing and how these affect learning. The selected battery aimed to provide us with an overall baseline of some of the cognitive processes important for learning and wellbeing, but within the limitations imposed by the wider remit of the study. Note This was the first UK study with a comparison group to collect a wide range of data from long term survivors of perinatally acquired HIV infection.

The decision of combining some of the included tests was based more on functions (e.g. immediate or delayed recall) than specific underlying neurocognitive concepts. However, acknowledging the relevance of the points made by the reviewer, we have included separate scores of each of the tests on the S1 table. The table would provide the reader with the opportunity of exploring performance of participants on each test within each study group.

We grouped the tests into domains based on the rationale alluded above, and used Cronbach α (a measure of internal consistency) to check these assumptions. We have clarified this in the methods section.

4. Although the results for the Verbal Learning PHIV C vs PHIV-C are statistically different, the confidence intervals are completely overlapping. Moreover, the z-scores are very similar (1.3 vs 1.5). This translates to a standard score of 78 vs 81, respectively. Is this really clinically meaningful? These scores may result in very similar behavioral presentation. Effect sizes should be reported to demonstrate magnitude of effects.

Response: The results of the ANOVA analysis show that the average z-scores in the three HIV/CDC groups are statistically significantly different which is also found in the regression analysis where HIV/CDC group was a statistically significant predictor of verbal learning scores. In both analyses PHIV/C had the lowest or predicted the lowest scores, followed by PHIV/no C and then the HIV- group. We have amended the wording of our interpretation of the results to suggest a trend can be seen. We have also presented the test scores by HIV/CDC group in S1 Table and the results of the regression analysis in S2 Table.

5. More clarity is warranted with regard to the aim and methods of the regression analysis:

a. What is the goal of the regression analysis? If part of the primary objective of the paper is to determine what factors might predict poorer cognitive functioning in this group, that should be made explicit in the introduction/aims.

Response: This has been added to the introduction.

b. IQ/education are important predictors of cognitive functioning. Were these controlled for in the regression to determine what factors predict poor outcomes? If not, why?

Response: IQ was not collected as part of the AALPHI study. Highest qualification was collected but was poorly completed with lots of missing data and therefore was not included in the analysis.

c. The authors note that the predictors used in the regression can be found in Table 1. Table 1 has over 20 variables. Are all of these included in the regression? I would think the analysis would be underpowered for that many variables. Please clarify and provide justification for the predictors.

Response: The variables with a * asterisk in Table 1 were included as predictors in the overall regression analysis. We have clarified this in the methods section. The variables chosen were reviewed by the AALPHI Steering Committee and included important variables which have been highlighted in other studies.

d. An additional analysis with only PHIV was conducted. What is the rationale for this? Was this decided a priori or is this a sensitivity analysis? This needed clarification.

Response: This analysis was conducted to investigate the effect of HIV-related factors on cognitive performance and was decided a priori. This has been clarified in the introduction and methods section.

6. Raw score means, SDs, and ranges for all cognitive testing should be presented. This is necessary for interpretation of the results.

Response: These data have now been included in a Supporting Information file (S1 Table).

Minor Revisions by Section

Abstract

1. The title and Abstract background make it sound like only learning and memory was assessed. Consider updating both.

Response: We thank the reviewer for their comment. As the main finding is related to these domains, we think the focus of the title is appropriate. The methods section of the abstract confirms the wider assessments conducted.

2. Acronyms are not defined before use. (e.g., CDC C diagnosis, ART)

Response: Thank you for your comment. They are now defined.

3. The original aim is to compare PHIV to HIV-PV, but there is a three-group comparison reported in results.

Response: We have amended the abstract and introduction to better reflect the three groups compared in the analysis.

4. Results of the linear regressions are not mentioned.

Response: We have included the regression analysis in the abstract and added it as an aim in the introduction.

5. The conclusion made in the abstract that results suggest earlier ART leads to better outcome is not support by the results reported in the abstract. There is no mention of ART in the result section.

Response: The reviewer is right in their observation. Our analysis did not show an association with ART exposure or age of ART initiation. However, our analysis clearly demonstrated previous AIDS-defining conditions differentiated the level of cognitive impairment in our study population. Since the main aim of ART is restoring or preventing deterioration of immune function and hence, preventing opportunistic infections or other AIDS-defining conditions; AIDS-defining conditions can be considered as a surrogate marker for delayed ART initiation which may or may not be due to late diagnosis with HIV infection. Participants with AIDS-defining conditions were mainly started on ART when presenting with these complications. Those participants who were started on ART before their immune functions was severely deteriorated showed better performance in all neuropsychological tests and, taking these two points together, we believe the conclusion is well supported. We have clarified our conclusions in the abstract and discussion.

Introduction

6. The connection to education in the introduction is a little unclear. This might be an important implication of the work; however, it is not directly related to the study aim.

Response: We decided to focus on executive function, verbal learning, and memory because these may impact outcomes in education and employability. We have amended the introduction to reflect this.

7. A rationale should be provided for using the HIV-negative control group. I assume this control group accounts for environmental and psychosocial factors related to HIV exposure. This should be made explicit.

Response: This rationale has been added to the methods section, where we think it fits better.

8. In line with the above, there is clearly good reason for comparing groups with and without a CDC C diagnosis separately, but the rationale should be made explicit.

Response: We have amended the introduction to making the inclusion of all three groups clearer.

9. Why are no a prioiri hypotheses reported?

Response: We have clarified the aims of the analysis in the introduction.

Method

10. The first sentence needs editing. It reads like the study was evaluating the impact of HIV infection and ART on HIV- young people as it is currently written.

Response: This has been amended.

11. Please report any exclusion criteria for the larger study.

Response: This has been added.

12. A lot of space is used to discuss the consent/assent process. This is good/ethical practice, but unnecessarily detailed for a manuscript.

Response: Some of the detail has been removed from this section.

13. Cognitive testing described to have occurred during a face-to-face interview. It may be better described as an interview with cognitive testing.

Response: This has been reworded.

14. Please define the CDC Clinical classification class C past history of an AIDS defining illness.

Response: The CDC C classification includes over 20 conditions. We have referenced the paper listing the conditions (apologies an incorrect reference was used in the original manuscript and has now been updated). If the editor would like all the conditions listed in the manuscript, then we can add them.

Statistical analysis

15. Specify what comparisons are being made using chi square tests/Wilcoxon/KW.

Response: This has been clarified.

Results

16. Were there any significant differences between your groups on important demographic or clinical factors such as education, SES, psychiatric diagnoses? I believe these comparisons are made in Table 1, but the authors should include a statement in the results and then point readers to the table.

Response: A sentence describing the significant differences has been added to the results section and Table 1 referenced.

17. Information on variable distribution/assumptions underlying statistical analyses should be reported.

Response: The normality of the data was assessed using visual plots and Sharpio-Wilk tests. Fisher’s exact tests for proportions were used if less than 80% of the cells had an expected value of five or greater. This information has been added to the Methods section.

Discussion

18. It would be interesting to discuss in more detail why your control group also performed lower than the normative sample on several tests to contextualize the results. For example, what common factors between all three groups could have impacted cognitive performance on these measures and is this consistent with the literature?

Response: Young people recruited as part of the comparison group were mainly siblings of PHIV or had a parent living with HIV and therefore, have been affected by HIV since their early childhood, making them comparable to our PHIV participants as they were likely to be exposed to all unmeasured factors that can affect development of cognitive functions. For instance, the majority of youngsters in both groups will have had a mother who was HIV positive and parental ill health, both physical and mental, and death and loss is much higher than in general population. Inconsistency or changes in care and the emotional and economic consequences of living with chronic parental ill health as well as the anxiety and confusion and uncertainty associated with managing a secret and stigmatising family diagnosis and add to the obstacles faced by many of these youngsters. These early chronic factors affect later learning and psychological outcomes and may explain the finding that scores on many tests fell below the published test norms in all study groups.

We have added more details about the HIV- comparison group to the methods section.

19. Based on what is reported in the discussion, the HIV+ cohort in this study had impairments in cognitive domains that differed from previously published worked. This should be discussed.

Response: Our study reported performance on all cognitive function domains below the normative mean of each test, which is consistent with several previous studies. In the discussion, we speculated unmeasured factors, such as socioeconomic variables, might affect testing performance (we have added a recent study Mbewe et al JAIDS 2022) and stated the importance of including an appropriate comparison group (i.e. HIV exposed but uninfected). Our results are also consistent with previous studies showing poorer performance in PHIV with previous AIDS-defining conditions when compared to those with less severe HIV disease and exposed but uninfected young people.

20. There is a note to add “blackmore reference” in the discussion.

Response: Thanks for pointing this out, it has been removed. The reference has already been included.

21. It is unclear why the second limitation is a limitation of the study. Was identifying differential effect of antiretroviral drugs on cognition a goal? Is there a reason to believe that different treatments affect cognition in different ways? If so, this should be made more explicit.

Response: Virological escape and compartmentalisation have been reported in PHIV. In patients treated with ART combinations with poorer penetration to the CNS or in those with other conditions affecting blood-brain barrier integrity, viral particles and consequent inflammatory mediators can be observed in CSF, even in patients with undetectable HIV-RNA in peripheral blood. Detectable HIV-RNA in CSF has been associated with a number of neurological negative outcomes, including impaired cognition. In addition, some anti-retroviral drugs may cause neuropsychiatric adverse effects some of which may interfere with neuropsychological testing. Therefore, not being able to explore the impact of ART exposure on cognition we believe is a limitation of the study.

22. A limitation of this study is only using active recall to test memory (i.e., no recognition memory test from HVLT), therefore it is unclear if deficits in delayed memory extend to recognition. If this cohort has impaired learning and recall, but not encoding (i.e., recognition memory is average), this would suggest a frontal executive, as opposed to amnestic mesial temporal profile.

Response: We agree with the reviewer, not including the verbal recognition component of the revised HVLT limits our ability of getting a full interpretation of the results if the aim of the analysis was understanding the neurological pathway leading to the observed performance level. However, the testing battery also included the NIH Toolbox Picture Sequence Test to explore episodic memory which measures episodic memory. Episodic memory could inform on medial temporal lobe function.

The neuropsychological test battery used for this analysis was one of several components included in the main study visit, which was decided in advance should not last more than two hours to make it more likely for each participant to complete the full assessment. That was the main reason for restricting the number of neuropsychological tests. In any case, the aim of the analysis conducted was to compare performance on specific cognitive function domains between PHIV and appropriately selected HIV negative young people.

23. The authors conclude that their findings support that earlier ART protects against cognitive development. Was age of ART included in the regression model? Did it predict outcomes? It seems you have the data to test this assumption, but it is not clearly reported anywhere in the manuscript.

Response: We agree with the reviewer (also see earlier response to minor revisions, comment 5), the tone of the conclusion seems to suggest ART initiation at an earlier age, may protect cognitive function. However, the intention was to state ART initiation at an earlier stage of the HIV disease (i.e. before it reaches severe immunosuppression and opportunistic diseases occur) would reduce the risk of cognitive impairment. The text of the manuscript was amended to reflect the basis for the conclusion. Age of ART initiation was included as a predictor in the PHIV only model, but it was not associated with cognitive performance. The methods section has been amended to clarify which variables were considered as predictors in the models.

Tables

24. Table 1 is very large. I would recommend splitting up into more than one table.

Response: The table has been split into Table 1 (Characteristics of all participants) and Table 2 (HIV-related factors for PHIV group).

25. Should present results of the linear regressions in a table for transparency.

Response: The results of the linear regressions have been added as Supplementary Information (S2 and S3 Tables).

Reviewer #2: Abstract

1. Background paragraph of abstract emphasizes relevance of learning and memory for successful educational and career progression, but this was not examined in this study. Authors should consider a more relevant background justification that would more directly apply to this work and supporting analyses.

Response: See response to reviewer 1, minor revisions, comment 6.

2. “CDC C” diagnosis term in abstract would not be known to a general scientific audience. Consider briefly defining.

Response: The definition of CDC C has been added to the abstract.

Methods

1. The section on the consent process can be shortened/streamlined.

Response: Some of the detail has been removed from this section.

2. It would be helpful for the authors to briefly explain why it would be relevant to compare PHIV/no C and PHIV/C. This is done within the discussion, but it would be beneficial for the reader to understand the reasoning for this comparison from the outset (perhaps within the introduction?)

Response: Our group and others have reported advanced HIV disease, defined as CDC stage C, to be an important risk factor for poorer cognitive function. In the same study population during their first neuropsychological assessment, participants in the PHIV/C performed worse than PHIV/no C and HIV- young people in some specific function domains, including executive function, speed of information processing and fine motor skills. Of note, at the initial assessment, we observed no difference between PHIV/no C and HIV- young people in any cognitive domain testing performance (Judd et al Clin Infect Dis 2016). We do agree with the reviewer, the rationale for comparing participants with different CDC status could be better explained on the text and we have amended the introduction accordingly.

3. Executive functioning and memory performance can be directly impacted by attention, and attention difficulties are documented in HIV. Was there a reason attention measures were not included in this battery?

Response: We do agree with the reviewer on the impact attention impairment would have on executive functioning and memory. Due to time constrains we were unable to include a more comprehensive neuropsychological testing battery and therefore, we decided to include tests that could explore additional cognitive function domains. For instance, as part of the tests used to explore executive functioning in our participants, we included the NIH Toolbox Flanker Inhibitory Control and Attention test that directly explores attention. So, if one wanted to explore the impact of attention impairment on the overall assessment of executive functioning, one could look into the Flanker Inhibitory Control test score. However, the aim of the analysis conducted was to compare performance on specific cognitive function domains, including executive functioning, between PHIV and appropriately selected HIV negative young people.

Results

1. While it is logical that multiple test scores within a domain would be averaged to present results in a cohesive manner, the authors should make explicit that the scores were sufficiently statistically similar for this to be appropriate. Please include a table summarizing mean performance per group for all individual tests using a common metric (e.g., z scores). Also explain the criteria for averaging across tests within a domain; for example, did the authors ensure that mean test scores within a domain were not statistically different from each other? I imagine that there may have been some variability in performance for within-domain tests. Is it possible that potential effects were attenuated because of variable performance within domains?

Response: Mean test scores and z-scores for each test have now been included in a Supporting Information file (S1 Table). As stated before, we grouped the tests into domains based on theoretical assumptions and used Cronbach α (a measure of internal consistency) to check these assumptions. We have clarified this in the methods section.

Discussion

1. There were some lower cognitive scores amongst participants of Black race found in adjusted multivariable models. It is important that the authors address potential explanations for this within the discussion.

Response: This follows a trend reported in our previous paper and we outline potential reasons for this trend in our previous paper. We have added a summary comment to the discussion of this paper.

General comment for consideration:

1. The term “Black” generally refers to a race/racial category, as you can have participants who are “Black” but report different ethnic backgrounds. The authors can consider these recently published suggestions for reporting of race or ethnicity: https://jamanetwork.com/journals/jama/fullarticle/2776936

Response: We have added to the methods section that ethnicity was self-reported, and clarified the categories used and which categories were combined for the purpose of the analysis. We have also capitalised the names of ethnicities. We have not reported detailed ethnicity breakdowns due to small numbers in some categories.

Submitted filename: Response to Reviewers v1.0.docx

Click here for additional data file.

27 Apr 2022

PONE-D-21-16526R1

Learning and memory function in young people with and without perinatal HIV in England

PLOS ONE

Dear Dr. Arenas-Pinto,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we have decided that your manuscript does not meet our criteria for publication and must therefore be rejected.

Specifically:

Your paper went out to one reviewer who assessed the revisions against the original reviews.  The reviewer noted that you had not satisfactorily addressed the initial concerns around situating the study in the literature, differentiating between this study and the previously published version of this study and that your grouping of tests required more justification. Given that the concerns are the same as the last time despite having a chance to address them I am afraid that the changes are just too onerous and I will have to reject the paper.

I am sure this news is not what you wanted to hear but I do not believe it would be fruitful to go back and forth on the same issues. I am sorry that we cannot be more positive on this occasion, but hope that you appreciate the reasons for this decision.

Kind regards,

Lori Buchanan, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: PONE-D-21-16526R1

The authors appropriately addressed many of the minor concerns and clarifications recommended by the reviewers; however, there are a few major concerns that were not addressed in this revision.

1. A more thorough review of the background literature would be helpful.

2. It is still not clear how this study adds to the previously published study. It sounds like the previous study included the same three groups and a variety of cognitive measures (presumably EF, learning, and memory measures). Is this a replication?

3. I still have concerns about the combination of neuropsychological tests scores. The authors have grouped the tests “based on theoretical assumptions”, but there is empirical evidence to suggest different groupings would make more sense based on the underlying neurocognitive functions assessed by these measures. The rationale for the groups needs to be strengthened.

4. The authors did not address the concern regarding the difference between statistical and clinical significance. Adding effect sizes would help readers understand the magnitude of effect for the statistically significant group difference.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

- - - - -

For journal use only: PONEDEC3

10 Jun 2022

Appeal request form - Response to Reviewer #1: PONE-D-21-16526R1

We provide a point-by-point response to the reviewer here. The reviewer’s concerns are in bold text and our response in italics.

As a way of background, in December 2021 we received numerous comments from two reviewers as well as a summary from the Academic Editor, in which the Editor noted that

“The reviewers and I agree that this work is of importance and that it should be published but both reviewers have highlighted areas where the paper can be strengthened to have a greater impact.”

This implies that the manuscript was already publishable but that improvements could lead to a stronger paper. We responded robustly to all of the comments, and made significant changes to the manuscript, although in hindsight could have improved the introduction further.

The authors appropriately addressed many of the minor concerns and clarifications recommended by the reviewers; however, there are a few major concerns that were not addressed in this revision.

1. A more thorough review of the background literature would be helpful.

In our last revision, we added more details to the introduction about our previous study looking at cognitive function at interview 1.

We have now added references and text to contextualise our study within recent literature. See tracked changes attached.

2. It is still not clear how this study adds to the previously published study. It sounds like the previous study included the same three groups and a variety of cognitive measures (presumably EF, learning, and memory measures). Is this a replication?

We actually wanted to measure for a second time performance in some of the cognitive functions we explored in interview 1, which were most affected in interview 1, but we decided not to look at change over time on any of these domains because we were using different tests. We decided to use different tests to increase comparability of our cohort to a similar study in the USA.

In our previous response to reviewer comments we said:

“As the reviewer mentioned, we have explored cognitive function performance in the AALPHI cohort at baseline. In the initial assessment we found no overall difference in neuropsychological testing performance between young people living with HIV, who have never developed any AIDS-defining complication, and HIV-negative young people. However, performance on tests exploring memory and information processing cognitive domains seem to be poorer in PHIV, particularly those who have experienced AIDS-defining complications in the past. Therefore, the round of neuropsychological assessment we are presenting in this manuscript, focused predominately on learning and memory cognitive function domains because of their implications on educational and employment outcomes. We have amended the introduction to reflect this.”

As stated above, we have already reported some differences in cognitive testing performance between participants with previous AIDS-defining conditions compared to those living with HIV but that have never progressed to CDC stage C conditions. The aim of this second neuropsychological assessment was to further explore cognitive domains considered particularly relevant to school progression and transition to adulthood. The assessment conducted is NOT a replication of our previous analysis.

We therefore have made no further changes.

3. I still have concerns about the combination of neuropsychological tests scores. The authors have grouped the tests “based on theoretical assumptions”, but there is empirical evidence to suggest different groupings would make more sense based on the underlying neurocognitive functions assessed by these measures. The rationale for the groups needs to be strengthened.

We believe that the reviewer has strong feelings against grouping tests in a way different to what they think is appropriate. We do not think it should be up to the reviewer to decide what rationale for combining tests is valid or not, but despite our explanations on the matter, they still seem unhappy.

To clarify, we used more than one test to explore three of the cognitive function domains evaluated. When exploring Executive Functioning, we used two NIH ToolBox tests that are designed and validated for that use (i.e. flanker inhibitory control and attention test and the dimensional change card sort test. The latest one, specifically designed and validated to explore cognitive flexibility).

When exploring language-related functions, we used the Weschler Individual Achievement Test® and the British Picture Vocabulary Scale, both commonly used and validated tests for this purpose.

Finally, because focusing on memory and learning skills was our main aim, we did conduct a number of different tests exploring that function domain. However, precisely because we wanted to investigate performance on each one of the components or interrelated specific tasks, we reported separately scores on verbal learning, verbal recall (HVLT-R) and episodic memory (NIH ToolBox: list sorting and picture sequence test). We also reported performance on attention and information processing speed separately.

Therefore, we seriously struggle to understand the objections the reviewer may have on our approach to report performance by cognitive function domain. We have however amended the methods section to reflect the rationale for grouping neuropsychological tests (lines 125 and 126 of the amended draft).

We previously responded to this comment with the following:

“We do agree that the findings do not clarify the specific neurocognitive processes most affected by the presence of HIV in the developing brain. The choice of tests recognised the complex interaction between attention, memory, information processing and how these affect learning. The selected battery aimed to provide us with an overall baseline of some of the cognitive processes important for learning and wellbeing, but within the limitations imposed by the wider remit of the study. Note This was the first UK study with a comparison group to collect a wide range of data from long term survivors of perinatally acquired HIV infection. The decision of combining some of the included tests was based more on functions (e.g. immediate or delayed recall) than specific underlying neurocognitive concepts. However, acknowledging the relevance of the points made by the reviewer, we have included separate scores of each of the tests on the S1 table. The table would provide the reader with the opportunity of exploring performance of participants on each test within each study group. We grouped the tests into domains based on the rationale alluded above, and used Cronbach α (a measure of internal consistency) to check these assumptions. We have clarified this in the methods section.”

4. The authors did not address the concern regarding the difference between statistical and clinical significance. Adding effect sizes would help readers understand the magnitude of effect for the statistically significant group difference.

We did not see any concerns regarding the difference between statistical and clinical significance in the previous set of reviewer comments. We responded to the request to add effect sizes in our previous response. We added further information on significant differences, and we added a supplementary table showing separate scores of each test. The discussion of the results does not include any speculation on the clinical impact of the differences we observed between the study groups on specific neuropsychological testing scores. Rather, we provided the reader with the scores by test and cognitive functional domain. Exploring the clinical impact of the differences observed was beyond the scope of this analysis and therefore, we have not made any further changes here.

Submitted filename: Justification for response to reviewers (v1.0; 18MAY2022).docx

Click here for additional data file.

28 Jul 2022

Submitted filename: PlosOneReview.docx

Click here for additional data file.

11 Aug 2022

1. The introduction and background review has been improved.

We thank the reviewer for their comment.

2. I am not doubting that the study adds to the literature. My confusion on what this study adds to their previous work comes from the lack of specificity in Line 70-77 (bolded).

“We have previously reported similar performance in neuropsychological testing in young people living with HIV, without a prior history of AIDS-defining complications, and a comparable group of HIV negative (HIV-) young people (YP) within the Adolescents and Adults Living with Perinatal HIV (AALPHI) cohort study (7). Of note, PHIV with a past history of AIDS74

defining conditions showed poorer performance in all cognitive function domains explored. In this paper we report on further cognitive testing performance of young people with PHIV

(with or without a previous AIDS-defining illness) and a comparable 76 group of HIV- young people.”

The difference between their first and second study is made clearer in their response to reviewers, but the rationale has still not been added to the introduction for readers. If I am interpreting correctly, the authors were (1) further clarifying the differences in memory and information processing found in previous work and (2) used tests more widely used in the literature to compare to other findings. If I am interpreting this correctly, the authors could add this rationale into the manuscript to increase clarity for readers.

We thank the reviewer for their suggestion. We have amended the introduction to address the point.

3. There is no one valid way to combine neuropsychological test scores. I provided empirical literature regarding the underlying neurocognitive functions that those tests measure. I did not state the authors must group tests as I suggested, only that it should be considered. For example, “verbal delayed recall” and “memory” tests are reported separately, but verbal delayed recall is memory. The authors show that the mean and Z-scores for the combined tests are comparable, which is helpful. Perhaps changing the name of the categories to better reflect the function they are attempting to assess would help.

We agree with the reviewer’s point. Verbal learning and verbal delayed recall are functions related to the memory domain. However, we consider cognitive functions related to management of verbal language particularly relevant to school progression and employability. Therefore, in addition to including tests to explore verbal application, we also considered important to provide the readers with the opportunity of looking at memory functions related to the used of verbal language.

We provide the scores of each individual test and hence, the reader has the possibility of observing performance on every function domain and every test explored.

4. Including the raw cognitive test data was very helpful. However, adding standardized effect sizes along with statistical significance testing demonstrates the magnitude of group difference and is common practice.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3444174/#:~:text=Effect%20size%20helps%20readers%20understand,the%20Abstract%20and%20Results%20sections.

We agree with the reviewer and have added standardized effect sizes to table S1; please see additional rows entitled “standardised score” for each domain/ test.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

15 Aug 2022

Learning and memory function in young people with and without perinatal HIV in England

PONE-D-21-16526R3

Dear Dr. Arenas-Pinto,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Lori Buchanan, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

thanks for your careful and responsive treatment of reviewers' comments.

Reviewers' comments:

6 Sep 2022

PONE-D-21-16526R3

Learning and memory function in young people with and without perinatal HIV in England

Dear Dr. Arenas-Pinto:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Lori Buchanan

Academic Editor

PLOS ONE