The IMPACT study: early loss of skeletal muscle mass in advanced pancreatic cancer patients.

BACKGROUND: Pancreatic cancer (PC) patients have multiple risk factors for sarcopenia and loss of skeletal muscle mass (LSMM), which may cause greater treatment toxicities, reduced response to cancer therapy, prolonged hospitalization, impaired quality of life, and worse prognosis.
METHODS: This is a retrospective study on advanced PC patients treated at the Department of Oncology of Udine, Italy, from January 2012 to November 2017. Among 162 patients who received chemotherapy, 94 consecutive patients with an available computed tomography (CT) scan were retrospectively analyzed. The primary objective of our study was to explore if an early LSMM ≥ 10% (measured at first radiological evaluation and compared with baseline) and/or baseline sarcopenia may impact prognosis. Baseline sarcopenia was defined according to Prado's criteria. Skeletal muscle area was measured as cross-sectional areas (cm2 ) using CT scan data through the Picture archiving and communication system (PACS) image system.
RESULTS: In the whole cohort, 48% of patients were ≤70 years old, and 50% had metastatic disease. At baseline, 73% of patients had sarcopenia, and 16% presented a visceral fat area ≥ 44 cm2 /m2 . Overall, 21% experienced an early LSMM ≥ 10%. Approximately 33% of sarcopenic patients at baseline and ~35% of patients with early LSMM ≥ 10% had a body mass index > 25 kg/m2 . Of note, 71% of patients were evaluated by a nutritionist, and 56% received a dietary supplementation (oral and/or parenteral). After a median follow-up of 30.44 months, median overall survival (OS) was 11.28 months, whereas median progression-free survival (PFS) was 5.72 months. By multivariate analysis, early LSMM ≥ 10% was significantly associated with worse OS [hazard ratio (HR): 2.16; 95% confidence interval (CI) 1.23-3.78; P = 0.007] and PFS (HR: 2.31; 95% CI 1.30-4.09; P = 0.004). Moreover, an exploratory analysis showed that inflammatory indexes, such as neutrophil-lymphocyte ratio variation, impact early LSMM ≥ 10% (odds ratio 1.31, 95% CI 1.06-1.61, P = 0.010).
CONCLUSIONS: Early LSMM ≥ 10% has a negative prognostic role in advanced PC patients. Further prospective investigations are needed to confirm these preliminary data.

Introduction

Pancreatic cancer (PC) is a very rapidly progressive disease characterized by several genetic and molecular alterations, as well as poor prognosis. Despite being a relatively rare type of cancer, recent estimates predict an increase in its incidence over the next years.1 Unfortunately, effective therapies capable of changing the disease's natural history are still lacking. PC patients have shorter survival and an increased risk of distant metastases.2, 3 Furthermore, the prognosis of these patients is conditioned by a higher incidence of malnutrition or cachexia, present in 70–80% of PC patients, responsible for at least 20% of all deaths.4, 5, 6, 7 In fact, multiple risk factors for loss of skeletal muscle mass (LSMM) due to cancer‐related factors and medical treatment concur to cause more treatment toxicities, asthenia, fatigue, reduced response to cancer therapy, prolonged hospitalization, impaired quality of life, and, therefore, a worse prognosis.2, 6, 8, 9 However, the identification of patients with muscle loss has become increasingly difficult, because ~40–60% of cancer patients are overweight or obese, even in the metastatic setting.3 Sarcopenia was defined by Evans as LSMM, which results in decreased strength and aerobic capacity and, thus, functional capacity. The pathogenesis of sarcopenia includes a systemic inflammatory reaction, involving anabolic and catabolic pathways responsible for skeletal muscle physiology.10 Firstly, inflammatory mediators such as cytokines [tumour necrosis factor (TNF‐α), interleukins (ILs)] and C‐reactive protein are released from the liver into the bloodstream, increasing the lipid and protein catabolism and, therefore, inhibiting the anabolic pathways. Moreover, they act on the central nervous system and visceral fat [visceral adipose tissue (VAT)], causing anorexia and fatigue.11

The best way to diagnose sarcopenia is by measuring the lean mass by either dual‐energy x‐ray absorptiometry (DXA) or computed tomography (CT) scan. Although DXA scans produce highly accurate results, they lack the ability to discriminate among lean and adipose tissue sub‐compartments. Conversely, CT image analysis at the third lumbar vertebra allows to distinguish adipose tissue (including visceral, subcutaneous, and intramuscular) from muscle tissue,6, 7, 8 and it has been validated as a reliable method for whole‐body composition assessment.

Recently, the use of gold standard methods for body composition assessment, including CT, has simplified the diagnosis of cancer‐related sarcopenia, thus better defining its prevalence. However, the impact survival of cachexia, including weight loss and/or sarcopenia, in PC patients has still been poorly studied.

Based on these premises, the aim of this study was to evaluate the prognostic impact of body composition among PC patients.

Material and methods

Study design

This is an observational, retrospective study that examined data of 165 advanced PC patients treated at the Oncology Department of Udine, Italy, from January 2012 to November 2017. A cohort of 94 consecutive patients with the availability of CT scan has been analysed. All patients had confirmed PC and the consent to the use of clinical data, rendered anonymous, for purposes of clinical research, epidemiology, training, and study of diseases for patients who have died and informed consent of the study for patients who are alive. The study was approved by the Departmental Review Board and by the Ethics Committee (approved in May 2018, Protocol Number 16662). Data have been obtained from electronic medical records according to strict privacy standards. The study aimed to characterize the prognostic impact of LSMM in advanced PC patients during first‐line treatment (i.e. baseline and after 12 weeks) analysing clinical outcome. Moreover, we evaluated the association among LSMM and inflammation index [neutrophil–lymphocyte ratio (NLR), lactate dehydrogenase (LDH), albumin] and with early dietary supplementation (defined as dietary implementation in the first 3 months of first‐line chemotherapy). Progression‐free survival (PFS) was defined as time from first‐line chemotherapy until progression of disease or death from any cause. Overall survival (OS) was defined as the time between treatment start and death from any cause.

Skeletal muscle mass assessment

Computed tomography scans, performed within 30 days since first‐line treatment start, were retrospectively evaluated. Skeletal muscle area was measured as a cross‐sectional area in cm2 using routine CT scan data through the PACS image system. Skeletal muscle index (SMI) was calculated as a cross‐sectional area of muscle (cm2) at the L3 level divided by the square of the height (m2). In particular, psoas, erector spinae, quadratus lumborum, transversus abdominis, external and internal oblique muscles of the abdomen, and rectus abdominis muscle were evaluated. Sarcopenia has been defined according to Prado's criteria [SMI < 53 cm2/m2 for men with body mass index (BMI) > 25, SMI < 43 cm2/m2 for men with BMI < 25, SMI < 41 cm2/m2 for women],2, 12, 13, 14 and early LSMM during chemotherapy has been defined as a decrease of SMI > 10% at first CT evaluation (approximately after 3 months).15

Blood sample analysis

Lactate dehydrogenase, albumin and C‐reactive protein levels were classified according to a cut‐off of 480 U/L, 35 g/L, and 5 mg/L, respectively. NLR was defined as the absolute neutrophil count divided by the absolute lymphocyte count. Full blood count data were eligible for analysis if performed within 1 month before the start of first‐line chemotherapy.

Statistical analysis

Patients' clinico‐pathological characteristics were summarized through descriptive analysis. Categorical variables were described through frequency distribution, whereas continuous variables will be reported through median range. Clinico‐pathological variables were compared between sarcopenic and non‐sarcopenic patients, and between patients who had an LSMM ≥ 10% and <10%.

The association of early LSMM with inflammation index was explored by Wilcoxon rank‐sum test or Kruskal–Wallis test, as statistically appropriate. The impact of clinico‐pathological features determining early LSMM and early nutritional intervention was analysed with logistic regression. Prognostic factors in terms of OS were tested in both univariate and multivariate models by Cox regression with 95% CI. Differences in survival were tested by log‐rank test and represented by Kaplan–Meier survival curves.

A two‐sided P < 0.05 was considered statistically significant. Statistical analysis was performed using STATA [StataCorp (2015) Stata Statistical Software: Release 14.2, College Station, TX, USA: StataCorp LP].

Results

This study included 94 patients with a diagnosis of advanced PC (patients' characteristics are shown in Table 1). In the whole cohort, 48% of patients were younger than 70 years, and 32% of patients had an Eastern Cooperative Oncology Group performance status (PS) of 0, and 50% had metastatic disease. Approximately 88% of patients had tumour histology or tumour cytology available showing a diagnosis of adenocarcinoma. The remaining 12% presented with clinical, radiological, and/or bio‐humoral features suggestive of pancreatic adenocarcinoma. Only 2% of patients received radiotherapy in multiple fractions. In addition, most of the patients (70%) did not receive surgery.

Table 1

Patients' characteristics

Variable	No. (n = 94)	%
Age
<70 years	45	48
≥70 years	49	52
Gender
Male	52	55
Female	42	45
Site of primary tumour
Head	51	54
Body–tail	43	46
ECOG PS at first‐line chemotherapy start
0	30	32
≥1	64	68
Stage at first‐line chemotherapy start
Locally advanced	47	50
Metastatic disease	47	50
Metastatic site at first‐line chemotherapy start
Bone	2	2
Liver	43	46
Lymph nodes	21	22
Lung	12	13
Peritoneum	7	8
BMI (kg/m2)
18.5–25	60	64
>25	34	36
Visceral fat (cm3/h2)
<44	79	84
≥44	15	16
Sarcopenia at first‐line chemotherapy start
Yes	69	73
No	25	27
Nutritional assessment (since diagnosis)
Yes	67	71
No	27	29
Nutritional supplementation (since diagnosis)
Yes	53	56
No	41	44
Type of first‐line chemotherapy
Single agent	25	27
Gemcitabine
Capecitabine
Combination regimen	53	56
Gemcitabine plus nab‐paclitaxel
Gemcitabine plus cisplatin
Fluorouracil, irinotecan, and oxaliplatin
Other	16	17
Lines of therapy for metastatic disease
1	51	54
2	27	29
3	16	17
Laboratory parameter at first‐line chemotherapy start
LDH (UI/L)
≤480	60	64
>480	14	15
Missing	20	21
Albumin (g/L)
≤35	20	21
>35	63	67
Missing	11	12
C‐reactive protein (mg/L)
≤5	18	19
>5	40	43
Missing	36	38
CA 19‐9 (UI/L)
≤60	18	19
>60	74	79
Missing	2	2

ECOG, Eastern Cooperative Oncology Group.

Overall, after a median follow‐up of 30.44 months, median OS was 11.28 months (25th–75th percentiles: 4.90–22.06 months), whereas median PFS was 5.72 months (25th–75th percentiles: 2.79–9.6 months).

As for anthropometric characteristics, median body weight of the whole cohort was 68.5 kg and median BMI was 24.1 kg/m2. Overall, 16% of patients presented a VAT ≥ 44 cm2/m2 (defined as visceral area divided by the square of the height), and 73% were sarcopenic at baseline (characteristics of sarcopenic patients were displayed in Table 2). Overall, 21% of population experienced an early LSMM ≥ 10% at first radiological assessment. On the other hand, 5% of population experienced a weight loss ≥ 10%. Characteristics of patients experiencing an early LSMM ≥ 10% are shown in Table 2. Among sarcopenic individuals, 12 patients experienced an early LSMM ≥ 10%. Approximately 33% of sarcopenic individuals at baseline and ~35% of patients with early LSMM had a BMI > 25 kg/m2.

Table 2

Characteristic of sarcopenic patients and patients with Δ SMI ≥ 10%

Variable	Sarcopenic patients n = 69		Patients with Δ SMI ≥ 10% n = 20
	No.	%	No.	%
Age
<70 years	30	43	10	50
≥70 years	39	57	10	50
Gender
Male	39	57	11	55
Female	30	43	9	45
Site of primary tumour
Head	38	55	8	40
Body–tail	31	45	12	60
Previous surgery
No	50	72	—	—
Yes	19	28
Diabetes
No	51	74	—	—
Yes	18	26
ECOG PS at first‐line chemotherapy start
0	20	29	6	30
≥1	49	71	14	70
Stage at first‐line chemotherapy start
Locally advanced	32	46	11	55
Metastatic disease	37	54	9	45
Metastatic site at first‐line chemotherapy start
Bone	2	3	1	5
Liver	34	49	9	45
Lymph nodes	15	21	6	25
Lung	10	14	3	15
Peritoneum	5	7	1	5
BMI (kg/m2)
18.5–25	46	67	13	65
>25	23	33	7	35
Visceral fat (cm3/h2)
<44	59	86	17	85
≥44	10	14	3	15
Nutritional assessment (since diagnosis)
Yes	50	72	17	85
No	19	28	3	15
Nutritional supplementation (since diagnosis)
Yes	40	58	13	65
No	29	42	7	35
Type of first‐line chemotherapy
Single agent	20	29	5	25
Combination regimen	38	55	9	45
Other	11	16	6	30
Lines of therapy for metastatic disease
1	37	53	13	65
2	21	30	4	20
3	11	17	3	15
Laboratory parameter at first‐line chemotherapy start
LDH (UI/L)
≤480	41	59	17	85
>480	11	16	1	5
Missing	17	25	2	10
Albumin (g/L)
≤35	14	20	6	30
>35	48	70	14	70
Missing	7	10	—	—
C‐reactive protein (mg/L)
≤5	13	19	6	30
>5	32	46	6	30
Missing	24	35	8	40
CA 19‐9 (UI/L)
≤60	12	17	4	20
>60	55	80	14	70
Missing	2	3	2	10

Δ, from baseline to first radiological assessment; ECOG, Eastern Cooperative Oncology Group.

Through univariate and multivariate analyses, we found that metastatic disease at diagnosis, VAT ≥ 44 cm2/m2, and an early LSMM ≥ 10% were negative prognostic factors in terms of both OS and PFS (Tables 3 and 4). More specifically, the median OS was 8.78 months for patients with an early LSMM ≥ 10% vs. 12.85 months for patients with an early LSMM < 10% [hazard ratio (HR): 2.16; 95% CI 1.23–3.78; P = 0.007, Figure 1]. Furthermore, median PFS was 4.14 months for the first group vs. 7.27 months for the second group (HR: 2.31; 95% CI 1.30–4.09; P = 0.004; Figure 2).

Table 3

Univariate and multivariate analyses for overall survival

Variable	Univariate analysis			Multivariate analysis
	HR	95% CI	P	HR	95% CI	P
Age
≥70 vs. <70 years	1.40	0.89–2.20	0.141
Stage at diagnosis
Metastatic disease vs. locally advanced	2.11	1.36–3.28	0.001	2.25	1.35–3.76	0.002
Diabetes
Yes vs. no	1.05	0.65–1.69	0.838
ECOG PS
≥1 vs. 0	1.44	0.88–2.34	0.141
BMI
>25 vs. 18.5–25 kg/m2	1.00	0.97–1.57	0.973
Visceral fat
≥44 vs. <44 cm3/h2	2.19	1.23–3.91	0.008	2.77	1.31–5.87	0.008
Site of primary tumour
Body–tail vs. head	1.36	0.88–2.12	0.160
Δ SMI
≥10% vs. <10%	1.92	1.12–3.31	0.018	2.16	1.23–3.78	0.007
Δ Karnofsky PS
≥20% vs. <20%	1.87	1.02–3.44	0.042	1.18	0.52–2.69	0.686
Nutritional assessment
Yes vs. no	1.03	0.63–1.68	0.905
Nutritional supplementation
Yes vs. no	1.43	0.91–2.23	0.112
Δ Body weight
≥10% vs. <10%	2.46	0.97–6.25	0.057
Sarcopenia
Yes vs. no	1.16	0.69–1.94	0.565
Early nutritional assessment
Yes vs. no	1.23	0.79–1.92	0.342
Early nutritional supplementation
Yes vs. no	1.49	0.95–2.33	0.077

The bold values are variables statistically significant in univariate and multivariate analysis.

Δ, from baseline to first radiological assessment; ECOG, Eastern Cooperative Oncology Group.

Table 4

Univariate and multivariate analyses for progression‐free survival

Variable	Univariate analysis			Multivariate analysis
	HR	95% CI	P	HR	95% CI	P
Age
≥70 vs. <70 years	1.19	0.77–1.84	0.420
Stage at diagnosis
Metastatic disease vs. locally advanced	1.88	1.22–2.90	0.004	1.84	1.11–3.04	0.017
Diabetes
Yes vs. no	0.98	0.62–1.56	0.961
ECOG PS
≥1 vs. 0	1.40	0.89–2.22	0.143
BMI
>25 vs. 18.5–25 kg/m2	0.84	0.53–1.32	0.460
Site of primary tumour
Body–tail vs. head	1.06	0.69–1.63	0.762
Δ SMI
≥10% vs. <10%	1.82	1.06–3.13	0.029	2.31	1.30–4.09	0.004
Visceral fat
≥44 vs. <44 cm3/h2	2.45	1.34–4.49	0.003	2.98	1.41–6.28	0.004
Δ Karnofsky PS
≥20% vs. <20%	1.97	1.08–3.60	0.027	1.43	0.63–3.23	0.383
Nutritional assessment
Yes vs. no	0.77	0.48–1.24	0.294
Nutritional supplementation
Yes vs. no	1.06	0.68–1.63	0.791
Δ Body weight
≥10% vs. <10%	2.60	1.03–6.59	0.043

The bold values are variables statistically significant in univariate and multivariate analysis.

Δ, from baseline to first radiological assessment; ECOG, Eastern Cooperative Oncology Group.

Figure 1

Overall survival. CI, confidence interval; HR, hazard ratio; LSMM, loss of skeletal muscle mass.

Figure 2

Progression‐free survival. CI, confidence interval; HR, hazard ratio; LSMM, loss of skeletal muscle mass.

To explore possible parameters that could predict an early LSMM superior or inferior to 10%, we performed a logistic regression revealing an association with NLR variation [odds ratio (OR) 1.31, 95% CI 1.06–1.61, P = 0.010] and LDH variation (OR 1.01, 95% CI 1.00–1.01, P = 0.020) (Figure 3). No association was observed with C‐reactive protein levels, Lymphocyte‐to‐monocytes ratio (LMR) variation, CA 19‐9 variation, albumin variation, BMI variation, or a body weight change after 3 months since diagnosis.

Figure 3

Association of inflammatory indexes with early nutritional intervention. LSMM, loss of skeletal muscle mass.

Overall, 71% of patients were evaluated by a nutritionist, and 56% of patients received a dietary supplementation (oral and/or parenteral). To explore possible factors associated with an earlier supplementation, we found an association only with body weight variation (OR 1.11, 95% CI 1.01–1.22, P = 0.024). No association was observed with a BMI variation, PS Karnofsky variation, baseline BMI, SMI variation, site of primary tumour, or age (Table 5, Figure 4). Only body weight variation at first radiological assessment during chemotherapy showed an association with the decision to prescribe an early supplementation.

Table 5

Logistic regression—univariate for early dietary supplementation

Variable	Univariate analysis
	HR	95% CI	P
Δ: Body weight	1.10	1.01–1.18	0.035
≤10% vs. >10%
BMI at first‐line chemotherapy start (kg/m2)	0.74	0.53–1.04	0.091
>25 vs. ≤25
Δ SMI	0.53	0.16–1.66	0.277
≤10% vs. >10%
Site of primary tumour	0.99	0.43–2.31	0.996
Body–tail vs. head
Age	0.79	0.34–1.84	0.595
≥70 vs. <70 years
Δ Karnofsky PS	1.00	0.97–1.03	0.796
≤20% vs. >20%

The bold values are variables statistically significant in univariate and multivariate analysis.

Δ, from baseline to first radiological assessment.

Figure 4

Association of body weight variation with early nutritional intervention.

Discussion

Recent literature has focused on the prognostic impact of sarcopenia and its post‐operative role in respect of complication.1 However, most of these studies included patients who received surgery, and, unfortunately,7, 16, 17 few data are available for advanced PC patients.

In the present study, we examined a cohort of 94 advanced PC patients, exploring in a real‐world scenario the prognostic impact of baseline sarcopenia (at the diagnosis of advanced PC), and of an early LSMM ≥ 10% (after 3 months). Additional anthropometric measures were evaluated (BMI, weight loss, and baseline VAT).

Approximately 73% of patients were sarcopenic at baseline according to previous data,2, 6, 7 and most of them had normal BMI (median BMI: 24.1 kg/m2). Furthermore, 33% of sarcopenic patients and 35% of patients with early LSMM had a BMI > 25 kg/m2. Of note, sarcopenic obesity may be potentially misleading because sarcopenia is also present among normal and obese patients.2, 18, 19 Therefore, in these cases, a skeletal muscle wasting may be masked, leading to greater toxicity and potentially serious adverse events to anticancer treatment.5, 20

Therefore, the use of BMI only is not reliable for nutritional assessment, because LSMM could be independent from weight loss and patients with similar weight or BMI may have a significant difference in body composition.21, 22, 23, 24 In our cohort, 21% of PC patients experienced an early LSMM ≥ 10%. Among sarcopenic patients, ~21% were subject to further loss of muscle mass ≥ 10%.

By univariate analysis, metastases at diagnosis, early PS loss according Karnofsky ≥ 20%, VAT ≥ 44 cm2/m2, and early LSMM ≥ 10% had a prognostic role. In multivariate analysis, in addition to the stage of diagnosis, only early LSMM ≥ 10% and VAT ≥ 44 cm2/m2 was statistically relevant and associated with worse prognosis in term of OS and PFS.

More recently, a retrospective analysis conducted on 782 PC patients found no significant prognostic association with survival outcomes for muscle attenuation.25 The considerable number of patients with early‐stage disease included in the study, ~35% of the whole cohort, may explain the discrepancy with our results, considering the possible differences in prognosis, the potential metabolic differences between advanced and early muscle wasting, and the recovering from muscle wasting after curative surgery.

Moreover, in contrast to Danai and colleagues who defined muscle attenuation as the average Hounsfield attenuation of all pixels in the muscle area, we evaluated the early LSMM as a decrease of SMI > 10% at first CT evaluation analysed in advanced PC.25

Cachexia in cancer patients is a well‐established negative prognostic factor,2 and sarcopenia may be one of its components, despite that the two conditions are very different. Cachexia is defined as a body weight loss over 12 months of 5% or more26; conversely, sarcopenia defines metabolic disorders that include LSMM, atrophy, fatigue, and reduction of muscular strength. During this process, muscle fibres are replaced by fibrotic tissue, resulting in neuromuscular junction degeneration, increased levels of reactive oxygen species alterations, and altered muscle metabolism.27, 28 Nowadays, the best index to identify sarcopenic patients is still under debate, as well as the nutritional follow‐up.4

Many studies are trying to clarify the molecular pathways underlying muscle wasting. The effect of systemic inflammation on muscle wasting could play a crucial role, because pro‐inflammatory factors such as cytokines (TNF‐α, IL‐1, IL‐6, and interferon γ), acute phase protein (C‐reactive protein), and metabolic factors impact metabolism, contributing to LSMM.29

In our study, we exploratorily evaluated the association between the early LSMM and systemic inflammation indexes and their impact on early LSMM considering NLR, LDH, C‐reactive protein, and albumin variation at 3 months after first‐line treatment. We found a significant association with NLR variations and LDH variations, although the LDH variation was not clinically relevant. No association was observed with other inflammatory indexes or with body weight loss. Notably, markers of systemic inflammation, such as NLR, are correlated with high levels of cytokines resulting in modulation of several factors (IGF‐1/Akt/mTOR axis, FoxOs, NF‐κB, and MAPK) favouring a catabolic state.27 More specially, IGF‐1 stimulates protein synthesis and promotes muscle hypertrophy; Akt/protein kinase B and FoxO are responsible for muscle homeostasis by regulating cell metabolism, growth, and survival. However, the main inhibitor of muscle growth is myostatin. It is produced by skeletal myocytes and negatively impact muscle growth through the interaction with IGF‐1–Akt signalling, mediated by the transcription factors small mothers against decapentaplegic 2 and 3.10, 30 Moreover, TNF‐α and IL‐6 inhibit myocyte differentiation promoting muscle atrophy and insulin resistance, leading to muscle wasting and increasing lipid deposition in skeletal muscle tissue.11, 28 LSMM could contribute to local inflammation, leading to further breakdown and enhancing systemic inflammation.11, 29, 31 Previous studies have confirmed the association of many of these inflammatory biomarkers with sarcopenia.27, 32, 33 In our study, we exploratorily evaluate the variation of these inflammatory indexes, but some of them were not associated with early LSMM, probably owing to the small sample size of the study and its retrospective nature.

In our cohort, 71% of patients received a nutritional assessment during any anticancer treatment line, and ~56% were evaluated in the first 3 months of first‐line chemotherapy. Moreover, 56% received a nutritional supplementation (oral and/or parenteral), and only 37% received an early nutritional intervention, although many of them were already sarcopenic. Overall, 85% of patients with early LSMM received a nutritional assessment, and only 55% received an early nutritional assessment. However, 30% received a dietary supplementation, and 25% received an early nutritional intervention (Table 2).

Surprisingly, early nutritional assessment and early dietary intervention did not impact prognosis in term of OS. Apparently, this finding was discordant with previous data.34, 35, 36 Therefore, we evaluated which factors were associated with early dietary supplementation, and only body weight loss was associated with an early dietary supplementation. However, as weight loss and BMI do not impact patient survival, nor does LSMM, it is useless to consider only these factors to perform a nutritional evaluation and implementation. Therefore, only a radiologic evaluation of LSMM provides an objective parameter that every clinician can use to monitor the nutritional status of PC patients.

Conclusions

Pancreatic cancer is a very rapidly progressive disease characterized by metabolic disorders that can easily lead to muscle mass depletion. The present study showed that an early LSMM ≥ 10% has a negative prognostic impact on advanced PC patients. On the other hand, baseline sarcopenia, weight loss, and BMI did not impact prognosis. Moreover, some inflammatory biomarkers, such as NLR variation, have an impact on an early LSMM ≥ 10%. Contrarily, early nutritional assessment and intervention were not associated with prognosis, and we showed that only body weight loss was associated with early nutritional implementation. Considering that weight loss and BMI did not impact patient survival, nor does LSMM, it is useless to consider only weight loss to perform a nutritional evaluation. In conclusion, to our knowledge, this is the first work to demonstrate the prognostic value of SMI loss during anticancer therapy independent from weight loss or gain. Therefore, to further evaluate the prognostic impact of an early nutritional treatment, we will provide a prospective Phase II study.

Funding

No funding was necessary for this work.

Conflict of Interest

The authors declare no competing financial interests.