Increased risk of anal squamous cell carcinoma in HIV-positive men with prior hepatitis B virus infection.

OBJECTIVE(S): HIV-positive individuals have elevated rates of anal squamous cell carcinoma (SCC), and sexually transmitted infections with its causative agent, high-risk human papillomavirus, and other oncoviruses including hepatitis B virus (HBV). HBV infection can cause liver cancer, and has been associated with increased risk of some extra-hepatic cancers including biliary tract cancer, pancreatic cancer, and non-Hodgkin lymphoma. Whether HBV is associated with anal SCC risk is unknown.
DESIGN: Prospective study of anal SCC risk in HIV-positive and HIV-negative MSM in the Multicenter AIDS Cohort Study from 1984 to 2014.
METHODS: Poisson regression models were used to examine the association between past or current HBV infection (positive tests for HBV core antibodies, surface antigen, and/or DNA) and anal SCC risk.
RESULTS: We observed 53 cases of anal SCC among 5298 participants with 79 334 person-years follow-up. Among HIV-positive men, past or current HBV infection was associated with anal SCC risk in models adjusted for age, CD4+ cell counts, HAART use, and other risk factors [incidence rate ratio (IRR), 95% confidence interval 3.15, 1.27-7.82]. Additional risk factors included immunological parameters 1 and 6 years prior to diagnosis (IRR, 95% confidence interval 2.45, 1.31-4.58 and 2.44, 1.3-4.59 for CD4+ cell counts <500 cells/μl; 2.43, 1.34-4.42 and 2.77, 1.5-5.11 for CD4+ : CD8+ ratios <0.5, respectively). Among HIV-negative men, IRR for prior HBV and anal SCC risk was similar, but NS due to small number of cases.
CONCLUSION: HIV-positive MSM with prior HBV infection have increased anal SCC risk. This population may benefit from screening.

Introduction

Hepatitis B virus (HBV) infection is a risk factor for hepatocellular carcinoma (HCC), and is also associated with elevated risk for some other types of cancer including pancreatic cancer [1-5] and non-Hodgkin lymphoma (NHL) [6-9]. The mechanisms by which HBV infection increases risk of some extra-hepatic cancers are unclear. However, its protumorigenic effects raise the possibility it could be an oncogenic cofactor [10-15].

HBV infection rates are high among HIV-infected populations and MSM [16-19]. HIV-infected populations have elevated risk for non-AIDS-defining malignancies (NADMs) linked to infections with oncoviruses such as HBV or hepatitis C (HCC) and high-risk human papilloma virus (hrHPV) (genital, anal, and oropharyngeal cancers) [20-31]. Anal squamous cell carcinoma (SCC) is a NADM caused by hrHPV in more than 90% of cases [32]. Anal hrHPV infections are detected in 30–90% of HIV-positive individuals, with the highest rates among MSM [32-37]. However, other factors are required for tumorigenesis [38]. Known risk for anal SCC factors include multiple sexual partners, immunodeficiencies, and smoking [39-42].

Synergistic syndemic interactions between hrHPV and HBV in promoting high-grade squamous intraepithelial lesions (HSIL) were suggested by a previous study [43]. Furthermore, anal hrHPV is more prevalent among individuals with prior HBV [44] or hepatitis [45]. Here, we evaluated the association between prior HBV infection and anal SCC risk in a prospective cohort of HIV-positive and HIV-negative MSM enrolled in the Multicenter AIDS Cohort Study (MACS).

Methods

Study cohort

This is a nested prospective cohort study in the MACS, an ongoing study established in 1984 that has enrolled 7343 HIV-positive and HIV-negative MSM. Clinical and laboratory data were collected at semiannual visits as described [25,39]. Eligible participants were 5298 MSM aged 30–70 years with at least 5 years of follow-up during 1984–2014 and one or more visits with HBV laboratory test data. Baseline was first visit after age 30 years. Institutional Review Boards at each study site approved the research and all participants provided written informed consent.

Data collection, risk factor classification, and cancer outcomes

The MACS public dataset release 25 was used for analyses. Past or current HBV infection was a time-invariant categorical variable defined by two or more positive lab tests for anti-HBV core antigen (ATHBC), HBV surface antigen (HBSAG), HBV e antigen (HBEAG), or HBV DNA (n = 2001), or single positive result for HBSAG (n = 29) or HBEAG (n = 1) any time following enrollment to study endpoint. Incident cancers were classified as described [25].

Statistical analysis

Participants were followed from baseline to first instance of incident cancer, last study visit, or age 70. Poisson regression models were restricted to HIV-positive participants. Factors associated with anal SCC in univariate analyses at P less than 0.2 were entered in the multivariate model, with stepwise backward selection used to retain significant features in the model (P < 0.05). Race and HAART use were retained in models as potential confounding factors. Models were sequentially adjusted for additional covariates having known associations with anal SCC risk and/or HBV. Statistical analysis was performed in R version 3.2.4 (R Project for Statistical Computing, Vienna, Austria).

Results

Demographic and behavioral characteristics by HIV and hepatitis B virus infection status

We identified 5298 MSM ages 30–70 enrolled in the MACS during 1984–2010 and contributing 79 334 person-years, with median follow-up of 12 years (Table 1). Thirty-eight percent (n = 2037) had past or current HBV infection at endpoint based on positive tests for ATHBC, HBSAG, HBEAG, and/or HBV DNA, of which 87.7% were positive at enrollment and 21% had one or more positive tests for HBSAG, HBEAG, and/or HBV DNA within follow-up. Age at endpoint, follow-up time, and frequencies of anal SCC and other NADMs were higher among participants with past or current HBV in comparison with HBV-negative participants. By contrast, frequencies of HIV-positive participants with ADMs or death were lower among participants with past or current HBV in comparison with HBV-negative participants. More HIV-negative participants with past or current HBV reported more than two sexual or anal receptive partners/6 months in comparison with HBV-negatives, whereas there was no difference by HBV status among only HIV-positive participants.

Table 1

Demographic and clinical characteristics of groups by HIV and HBV status.

	All		HIV-negative		HIV-positive
	HBV-negative, n = 3261	Past or current HBV, n = 2037	HBV-negative, n = 1725	Past or current HBV, n = 848	HBV-negative, n = 1536	Past or current HBV, n = 1189
Age at baseline (median [IQR])	34 [30, 39]	35 [30, 41]	34 [30, 40]	36 [31, 42]	34 [30, 38]	34 [30, 40]
Age at endpoint (median [IQR])	45 [39, 53]	56 [47, 63]	48 [41, 56]	60 [52, 66]	43 [38, 50]	53 [45, 60]
Cumulative person-years (median [IQR])	11 [7, 13]	20 [12, 28]	11 [9, 17]	24 [12, 30]	9 [6, 12]	14 [11, 27]
Race
White	2601 (79.8)	1476 (72.5)	1428 (82.8)	690 (81.4)	1173 (76.4)	786 (66.1)
African-American	379 (11.6)	394 (19.3)	188 (10.9)	109 (12.9)	191 (12.4)	285 (24.0)
Other	281 (8.6)	167 (8.2)	109 (6.3)	49 (5.8)	172 (11.2)	118 (9.9)
No. of sexual partners >2/6 monthsa	2159 (66.2)	1474 (72.4)	1036 (60.1)	616 (72.6)	1123 (73.1)	858 (72.2)
No. of anal receptive sexual partners >2/6 monthsa	684 (21.0)	515 (25.3)	150 (8.7)	120 (14.2)	534 (34.8)	395 (33.2)
Tobacco ≥0.5 packs/dayb	788 (24.2)	469 (23.0)	363 (21.0)	159 (18.8)	425 (27.7)	310 (26.1)
Alcohol >14 drinks/week or bingeingb	743 (22.8)	469 (23.0)	360 (20.9)	196 (23.1)	383 (24.9)	273 (23.0)
Marijuana ≥1000 exposuresc	252 (7.7)	180 (8.8)	88 (5.1)	56 (6.6)	164 (10.7)	124 (10.4)
Poppers ≥1 year of daily or weekly use	729 (22.4)	605 (29.7)	321 (18.6)	239 (28.2)	408 (26.6)	366 (30.8)
Crack cocaine ≥100 exposuresc	76 (2.3)	111 (5.4)	42 (2.4)	28 (3.3)	34 (2.2)	83 (7.0)
CD4+ cell count (cells/μl) (median [IQR])d	–	–	988 [765, 1250]	938 [734, 1183]	291 [88, 567]	489 [272, 703]
CD4+ cell count <350 cells/μld	–	–	6 (0.3)	9 (1.1)	855 (55.7)	399 (33.6)
CD4+ nadir (cells/μl) (median [IQR])e	–	–	640 [494, 813]	606 [475, 770]	157 [30, 352]	215 [95, 342]
CD4+ : CD8+ ratio (median [IQR])d	–	–	1.84 [1.37, 2.48]	1.79 [1.32, 2.35]	0.34 [0.13, 0.70]	0.57 [0.31, 0.88]
CD4+ : CD8+ ratio <1d	–	–	128 (7.4)	82 (9.7)	1319 (85.9)	964 (81.1)
CD4+ : CD8+ ratio nadir (median [IQR])e	–	–	1.22 [0.91, 1.61]	1.13 [0.89, 1.48]	0.20 [0.06, 0.41]	0.24 [0.12, 0.42]
HIV viral load >400 copies/mld	–	–	–	–	1048 (73.0)	389 (33.7)
HAART use at endpoint	–	–	–	–	494 (32.2)	911 (76.6)
HBV-active medication use prior to endpointf	375 (11.5)	755 (37.1)	7 (0.4)	30 (3.5)	368 (24.0)	725 (61.0)
HBV past or current
ATHBC positiveg	–	2003 (98.3)	–	829 (97.8)	–	1174 (98.7)
HBSAG, HBEAG, and HBV DNA negativeh	–	1610 (79.0)	–	707 (83.4)	–	903 (75.9)
HBSAG, HBEAG, or HBV DNA positive	–	393 (19.3)	–	122 (14.4)	–	271 (22.8)
ATHBC negative/HBSAG or HBV DNA positive	–	34 (1.7)	–	19 (2.2)	–	15 (1.3)
HCV past or current infectioni	256 (7.9)	273 (13.4)	67 (3.9)	73 (8.6)	189 (12.3)	200 (16.8)
ADMsj	434 (13.3)	152 (7.5)	8 (0.5)	8 (0.9)	426 (27.7)	144 (12.1)
NADMs
Anal SCC	8 (0.2)	45 (2.2)	2 (0.1)	6 (0.7)	6 (0.4)	39 (3.3)
Liver cancer	3 (0.1)	17 (0.8)	1 (0.1)	6 (0.7)	2 (0.1)	11 (0.9)
Other NADMsk	161 (4.9)	182 (8.9)	85 (4.9)	79 (9.3)	76 (4.9)	103 (8.7)
Deaths	445 (13.6)	181 (8.9)	33 (1.9)	22 (2.6)	412 (26.8)	159 (13.4)

Data shown are n (%) unless otherwise indicated. ADM, AIDS defining malignancy; ATHBC, antibody to HBV core antigen; HBEAG, HBV e antigen; HBSAG, HBV surface antigen; HBV, hepatitis B virus; HCV, hepatitis C virus; IQR, interquartile range; NADM, non-AIDS-defining malignancy; SCC, squamous cell carcinoma.

aAverage over first three visits.

bAverage over 10 years prior to endpoint.

cTotal exposures over 10 years prior to endpoint.

dTime-updated values lagged 1 year prior to endpoint.

eLowest value between enrollment and study endpoint.

fHBV-active medications included lamivudine (n = 952), tenofovir (n = 618), emtricitabine (n = 134), pegylated-interferon or IFN-α (n = 79), and other (n = 25) between enrollment and study endpoint.

gAt least two positive tests, or one positive test in combination with positive test for HBSAG, HBEAG, or HBV DNA any time following enrollment to study endpoint.

hNegative = 0 positive tests among subjects with at least 1 value.

iAt least one positive test for HCV antibodies or HCV RNA any time following enrollment to study endpoint.

jKaposi sarcoma and/or non-Hodgkin lymphoma diagnosed within study period.

kAll other NADMs (excluding nonmelanoma skin cancers and benign tumors) diagnosed within study period.

Immunological and virological characteristics by HIV and hepatitis B virus infection status

CD4+ cell counts and CD4+ : CD8+ ratios were lower in HIV-positive compared with HIV-negative participants, whereas the relationship of these variables to HBV status varied by HIV status (Table 1). Among HIV-negative participants with past or current HBV, CD4+ cell counts and CD4+ nadirs were lower at 1 year prior to endpoint in comparison with HBV-negative . Among HIV-positive participants with past or current HBV, CD4+ cell counts and CD4+ : CD8+ ratios were higher, fewer participants had CD4+ cell counts less than 350 cells/μl at 1 year prior to endpoint, and HAART or HBV-active medication use was more frequent in comparison with HBV-negative participants (Table 1).

Crude incidence of anal squamous cell carcinoma and other virus-associated cancers

We observed 53 incident cases of anal SCC during follow-up, eight in HIV-negative participants and 45 in HIV-positive participants. Crude incidence rates of anal SCC were higher in participants with past or current HBV regardless of HIV status (Supplemental Digital Content 1). Past or current HBV was associated with increased risk of anal SCC among HIV-positive participants [crude incidence rate ratio (IRR), 95% confidence interval (CI) 4.92, 2.09–11.63]. Similar IRR was observed among HIV-negative participants (IRR, 95% CI 4.05, 0.82–20.08), but NS due to a low number of cases. Crude incidence rates of liver cancer were higher in participants with vs. without past or current HBV regardless of HIV status. Incidence rates of other NADMs were similar by HBV status (IRRs 1.04–1.27), whereas incidence rates of Kaposi sarcoma and NHL were lower in HIV-positive participants with past or current HBV.

Hepatitis B virus and anal squamous cell carcinoma risk in HIV-positive subjects

In univariate models, past or current HBV was associated with increased risk of anal SCC in HIV-positive participants (IRR, 95% CI 4.92, 2.09–11.63), as was older age and CD4+ cell counts less than 500 or CD4+ : CD8+ ratios less than 0.5 lagged 6 years prior to endpoint (Supplemental Digital Content 2). The association remained significant in multivariate models adjusted for age, race, HAART use, and CD4+ cell counts or CD4+ : CD8+ ratios (IRRs, 95% CIs 3.07–3.29, 1.24–7.64 to 1.32–8.23; Table 2). In subanalyses of either past or current HBV compared with HBV-negative participants, only past HBV had a significant association with anal SCC risk (IRRs, 95% CIs 3.39–3.71, 1.32–8.75 to 1.42–9.67; Table 2). Further adjustment for HCV infection, number of sexual partners, tobacco use, or poppers use did not attenuate this association (Table 2). CD4+ cell counts less than 500 or CD4+ : CD8+ ratios less than 0.5 at 6 years prior to endpoint were associated with increased anal SCC risk in adjusted models (Table 2). CD4+ cell counts less than 500 or CD4+ : CD8+ ratios less than 0.5 at 1 year prior to endpoint were significant only in adjusted models, whereas nadir values were marginally significant (Supplemental Digital Contents 2 and 3 and Table 2). A negative association between no HAART use at endpoint and anal SCC risk was significant in unadjusted, but not adjusted models. Older age had a strong association with increased anal SCC risk in adjusted models, tobacco use had a marginally significant association (IRRs 1.60–1.83; P = 0.068–0.159) (Supplemental Digital Content 3), and race, number of sexual partners, HCV infection, or poppers use had no significant associations (Table 2).

Table 2

Multivariate analysis of risk factors associated with anal squamous cell carcinoma in HIV-positive participants.

	CD4+ variables lagged 1 year				CD4+ variables lagged 6 years
Multivariate models	IRR (95% CI)	P	IRR (95% CI)	P	IRR (95% CI)	P	IRR (95% CI)	P
Model 1
HBV past or current	3.26 (1.31, 8.16)	0.011	3.29 (1.32, 8.23)	0.011	3.15 (1.27, 7.82)	0.013	3.07 (1.24, 7.64)	0.016
Age group: 30–49	Reference		Reference		Reference		Reference
50–59	4.96 (2.52, 9.75)	<0.001	5.1 (2.59, 10.02)	<0.001	4.94 (2.51, 9.72)	<0.001	5.01 (2.55, 9.85)	<0.001
60–70	4.56 (1.81, 11.51)	0.001	4.78 (1.90, 12.05)	<0.001	4.57 (1.81, 11.53)	0.001	4.52 (1.79, 11.42)	0.001
Race: Non-African-American	Reference		Reference		Reference		Reference
African-American	0.99 (0.46, 2.12)	0.973	0.97 (0.45, 2.09)	0.94	0.97 (0.45, 2.09)	0.937	0.95 (0.44, 2.05)	0.905
No HAART use	0.56 (0.19, 1.72)	0.315	0.53 (0.17, 1.61)	0.259	0.71 (0.24, 2.13)	0.539	0.75 (0.25, 2.27)	0.613
CD4+ cell count <500 cells/μl	2.45 (1.31, 4.58)	0.005	–		2.44 (1.30, 4.59)	0.006	–
CD4+ : CD8+ ratio <0.5	–		2.43 (1.34, 4.42)	0.004	–		2.77 (1.50, 5.11)	0.001
Model 2
Past HBVa	3.59 (1.38, 9.35)	0.009	3.71 (1.42, 9.67)	0.007	3.49 (1.35, 9.01)	0.010	3.39 (1.32, 8.75)	0.011
Age group: 30–49	Reference		Reference		Reference		Reference
50–59	5.83 (2.77, 12.27)	<0.001	6.07 (2.88, 12.77)	<0.001	5.84 (2.77, 12.32)	<0.001	5.94 (2.82, 12.52)	<0.001
60–70	4.10 (1.39, 12.07)	0.010	4.37 (1.48, 12.85)	0.007	4.10 (1.39, 12.06)	0.010	4.10 (1.39, 12.07)	0.010
Race: Non-African-American	Reference		Reference		Reference		Reference
African-American	0.55 (0.19, 1.55)	0.258	0.54 (0.19, 1.53)	0.249	0.54 (0.19, 1.51)	0.240	0.52 (0.19, 1.48)	0.221
No HAART use	0.74 (0.23, 2.35)	0.611	0.66 (0.21, 2.11)	0.486	0.93 (0.30, 2.91)	0.902	0.96 (0.31, 3.01)	0.948
CD4+ cell count <500 cells/μl	2.17 (1.11, 4.23)	0.023	–		2.53 (1.28, 5.03)	0.008	–
CD4+ : CD8+ ratio <0.5	–		2.62 (1.36, 5.04)	0.004	–		2.90 (1.50, 5.62)	0.002
Model 3
Current HBVb	1.26 (0.56, 2.83)	0.576	1.19 (0.53, 2.69)	0.671	1.26 (0.56, 2.83)	0.576	1.17 (0.52, 2.64)	0.701
Age group: 30–49	Reference		Reference		Reference		Reference
50–59	5.40 (2.75, 10.61)	<0.001	5.55 (2.82, 10.90)	<0.001	5.39 (2.74, 10.59)	<0.001	5.48 (2.79, 10.77)	<0.001
60–70	5.25 (2.08, 13.24)	<0.001	5.47 (2.17, 13.79)	<0.001	5.23 (2.07, 13.18)	<0.001	5.18 (2.05, 13.06)	<0.001
Race: Non-African-American	Reference		Reference		Reference		Reference
African-American	1.04 (0.48, 2.23)	0.925	1.03 (0.48, 2.22)	0.937	1.03 (0.48, 2.21)	0.947	1.00 (0.47, 2.16)	0.992
No HAART use	0.35 (0.12, 1.01)	0.051	0.32 (0.11, 0.94)	0.039	0.45 (0.16, 1.31)	0.143	0.48 (0.17, 1.40)	0.180
CD4+ cell count <500 cells/μl	2.43 (1.30, 4.54)	0.006	–		2.46 (1.30, 4.62)	0.005	–
CD4+ : CD8+ ratio <0.5	–		2.37 (1.30, 4.34)	0.005	–		2.82 (1.53, 5.22)	<0.001
Model 1 + HCVc
HBV past or current	3.26 (1.31, 8.16)	0.011	3.29 (1.32, 8.22)	0.011	3.15 (1.27, 7.82)	0.013	3.07 (1.23, 7.63)	0.016
Model 1 + no. sexual partnersd
HBV past or current	3.26 (1.30, 8.16)	0.012	3.29 (1.32, 8.23)	0.011	3.15 (1.27, 7.83)	0.013	3.06 (1.23, 7.62)	0.016
Model 1 + tobaccoe
HBV past or current	3.19 (1.28, 7.98)	0.013	3.21 (1.29, 8.03)	0.012	3.09 (1.25, 7.66)	0.015	3.00 (1.21, 7.47)	0.018
Model 1 + poppersf
HBV past or current	3.45 (1.38, 8.62)	0.008	3.46 (1.39, 8.64)	0.008	3.33 (1.34, 8.27)	0.009	3.2 (1.29, 7.96)	0.012

Poisson regression models: P values calculated using Wald's test, bold indicates P less than 0.05. All HIV-positive participants are included (n = 2725). CI, confidence interval; HBEAG, HBV e antigen; HBSAG, HBV surface antigen; HBV, hepatitis B virus; HCV, hepatitis C virus; IRR, incidence rate ratio.

aParticipants with at least two positive ATHBC test results compared with HBV-negative participants.

bParticipants with at least one positive HBSAG, HBEAG, or HBV DNA test result compared with HBV-negative participants.

cPositive tests for HCV antibodies or HCV RNA prior to endpoint compared with participants with negative tests, restricted to participants with available tests (n = 2653).

dMore than two sexual partners/6 months compared with two partners or less/6 months, average over first three visits.

eAt least 0.5 packs/day compared with less than 0.5 packs/day, average over 10 years prior to endpoint. Supplemental Digital Content 3 shows full models including tobacco use.

fAt least 1 year of daily or weekly use compared with monthly, less than monthly, or no use within follow-up.

Demographic and clinical characteristics of groups by HIV status and anal squamous cell carcinoma diagnosis

We compared characteristics of the cohort stratified by HIV and anal SCC diagnosis (Supplemental Digital Content 4). Among HIV-positive participants, lab values indicating past HBV infection were more common among anal SCC cases compared with noncases. Similar trends were observed among HIV-negative participants. Anal SCC cases were older at endpoint, with longer follow-up and lower death rates in comparison with noncases among HIV-positive participants, anal SCC cases had lower CD4+ cell counts and CD4+ : CD8+ ratios lagged 6 years prior to endpoint among HIV-positive participants and lower nadir CD4+ cell counts and nadir CD4+ : CD8+ ratios compared with noncases in HIV-negative participants. HAART and HBV-active medication use were more common among anal SCC cases compared with noncases.

Discussion

In this prospective study of 5298 MSM enrolled in the MACS from 1984 to 2014, past or current HBV was associated with three-fold increased risk of anal SCC among HIV-positive participants, and remained an independent risk factor in models adjusted for age, HAART use, CD4+ cell counts or CD4+ : CD8+ ratios, and other risk factors. We also confirmed previously reported associations between anal SCC risk and older age, decreased CD4+ cell counts or CD4+ : CD8+ ratios, and smoking. Together with prior studies demonstrating increased HSIL among individuals coinfected with anal hrHPV and HBV [43], our findings suggest that HBV may be an oncogenic cofactor for development of anal SCC.

The association between HBV infection and anal SCC occurred primarily in HIV-positive participants with resolved HBV [46]. In analyses limited to active or chronic HBV, the association between HBV and anal SCC was NS. Similarly, Hassan et al.[1] found the association between HBV and increased risk of pancreatic cancer was evident in ATHBC-positive, HBSAG-negative individuals. These findings raise the interesting question of how resolved HBV infections might impact cancer risk later in life.

HBV has been associated with elevated risk of several nonhepatic cancers and NHL [1-9]. HBV integrates into host DNA, causing insertional mutagenesis, and encodes the oncogenic hepatitis B X protein, which targets tumor suppressor and DNA damage repair pathways [10,12,13,15]. With regard to indirect roles in promoting risk of anal SCC, HBV induces inflammation and T-cell exhaustion [11,14] and cirrhosis can lead to immune dysregulation and CD4+ lymphopenia [47-49]. However, immunological effects of resolved HBV infections are unknown. We found a significant association between low CD4+ cell counts or CD4+ : CD8+ ratios at 6 years prior to endpoint and anal SCC risk, consistent with [50]. Further studies are warranted to determine if resolved HBV is associated with immunological effects that impact later risk of cancers.

Participants on HAART had greater risk of anal SCC in comparison with those not on HAART, consistent with [39,50]. AIDS-related deaths are a competing risk and HAART prolongs life expectancy, allowing more time during which precursor lesions can progress to anal cancers [39,40,50-52].

We acknowledge limitations of our study. Lack of data related to hrHPV infection precluded modeling its role in anal SCC carcinogenesis. Our study was limited to mostly white MSM in a cohort at high risk of acquiring HIV, HBV, and HPV, with high rates of smoking and substance use, limiting applicability of our findings to the general population. HIV-positive participants with past or current HBV were slightly older, with more HAART and HBV-active medication use, better immunological profiles, and lower death rates. However, a case–control analysis in which each anal SCC case was matched to 10 controls based on recruitment wave, age at enrollment, race, HIV-infection status, smoking, follow-up, and calendar year at endpoint supported all of our main conclusions (data not shown).

In conclusion, this study demonstrates a significant association between prior HBV infection and increased anal SCC risk independent of age, HAART use, and immunological parameters. These results have implications regarding potential benefits of anal SCC screening in HIV-negative and HIV-positive MSM with prior HBV, and suggest that HBV vaccination in at-risk populations has the potential to improve anal cancer prevention.

Acknowledgements

Data for this article was obtained by the Multicenter AIDS Cohort Study (MACS) with centers at Baltimore (U01-AI35042): The Johns Hopkins University Bloomberg School of Public Health: Joseph B. Margolick (PI), Jay Bream, Todd Brown, Barbara Crain, Adrian Dobs, Richard Elion, Michelle Estrella, Lisette Johnson-Hill, Sean Leng, Anne Monroe, Cynthia Munro, Michael W. Plankey, Wendy Post, Ned Sacktor, Jennifer Schrack, Chloe Thio; Chicago (U01-AI35039): Feinberg School of Medicine, Northwestern University, and Cook County Bureau of Health Services: Steven Wolinsky (PI), John P. Phair, Sheila Badri, D.G., David Ostrow, Frank J. Palella, Jr., Sudhir Penugonda, Susheel Reddy, Matthew Stephens, Linda Teplin; Los Angeles (U01-AI35040): University of California, UCLA Schools of Public Health and Medicine: Roger Detels (PI), Otoniel Martinez-Maza (Co-PI), Aaron Aronow, Peter Anton, Robert Bolan, Elizabeth Breen, Anthony Butch, Shehnaz Hussain, Beth Jamieson, Eric N. Miller, John Oishi, Harry Vinters, Dorothy Wiley, Mallory Witt, Otto Yang, Stephen Young, Zuo Feng Zhang; Pittsburgh (U01-AI35041): University of Pittsburgh, Graduate School of Public Health: Charles R. Rinaldo (PI), Lawrence A. Kingsley (Co-PI), James T. Becker, Phalguni Gupta, Kenneth Ho, Susan Koletar, Jeremy J. Martinson, John W. Mellors, Anthony J. Silvestre, Ronald D. Stall; Data Coordinating Center (UM1-AI35043): The Johns Hopkins University Bloomberg School of Public Health: Lisa P. Jacobson (PI), Gypsyamber D'Souza (Co-PI), Alison Abraham, Keri Althoff, Jennifer Deal, Priya Duggal, Sabina Haberlen, Alvaro Munoz, Derek Ng, Janet Schollenberger, Eric C. Seaberg, Sol Su, Pamela Surkan. Website located at http://www.statepi.jhsph.edu/macs/macs.html.

Cancer incidence data were provided by the following state agencies: Maryland Cancer Registry, Center for Cancer Prevention and Control, Department of Health and Mental Hygiene, Baltimore, MD 21201; Illinois Department of Public Health, Illinois State Cancer Registry; Bureau of Health Statistics & Research, Pennsylvania Department of Health, Harrisburg, Pennsylvania; Ohio Cancer Incidence Surveillance System (OCISS), Ohio Department of Health (ODH), a cancer registry partially supported in the National Program of Cancer Registries at the Centers for Disease Control and Prevention (CDC) through Cooperative Agreement No. 5U58DP000795-05; and California Department of Public Health pursuant to California Health and Safety Code Section 103885; CDC's National Program of Cancer Registries, under cooperative agreement 5NU58DP003862-04/DP003862; the National Cancer Institute's Surveillance, Epidemiology and End Results Program under contract HHSN261201000140C awarded to the Cancer Prevention Institute of California, contract HHSN261201000035C awarded to the University of Southern California, and contract HHSN261201000034C awarded to the Public Health Institute. We acknowledge the State of Maryland, the Maryland Cigarette Restitution Fund, and the National Program of Cancer Registries of the CDC for the funds that support the collection and availability of the cancer registry data. The analyses, findings, interpretations, and conclusions of this report are those of the authors. No endorsement by any of the states providing data, the National Cancer Institute, the CDC, or their Contractors and Subcontractors is intended nor should be inferred. The authors acknowledge Steven Wolinsky, Sukrutha Chettimada, Debjani Guha, and Shibani Mukerji for helpful discussions and Elizabeth Carpelan for assistance with article preparation.

J.A. participated in study design, performed data assembly and statistical analysis, drafted the article, and prepared tables and figures. D.R.L. and V.M. participated in data parsing, assembly, and analysis. H.U. participated in study design and statistical analysis. D.G. conceived of the study, supervised its design, coordination, assembly, and analysis, and helped write and edit the article. All authors read, participated in editing the article, and approved the final article.

The work was supported by NIH grants to D.G. (R01 DA030985 and DA046203) and in part by NIH funding to the Northwestern University Clinical Research Unit of the MACS [U01-AI35039, with additional cofunding from National Cancer Institute (NCI) and National Institute on Drug Abuse (NIDA)]. J.A. was supported in part by NIH T32-AI007386-25. The MACS is funded by the National Institute of Allergy and Infectious Diseases (NIAID; U01-AI35039, U01-AI35040, U01-AI35041, U01-AI35042, and UM1-AI35043), with additional cofunding from the National Cancer Institute (NCI), National Institute on Drug Abuse (NIDA), and National Institute of Mental Health (NIMH) at the National Institutes of Health (NIH). MACS data collection is also supported by UL1-TR000424 (JHU CTSA).

Conflicts of interest

There are no conflicts of interest.