Impact on survival of sarcopenia, systemic inflammatory response and anthropometric factors after pancreatectomy for resectable pancreatic adenocarcinoma

Abstract

Introduction

Pancreatic adenocarcinoma (PDAC) is becoming a public health issue with a 5-years survival rate around 10%. Patients with PDAC are often sarcopenic, which impacts postoperative outcome. At the same time, overweight population is increasing and adipose tissue promotes tumor related-inflammation. With several studies supporting independently these data, we aimed to assess if they held an impact on survival when combined.

Methods

We included 232 patients from two university hospitals (CHU de Lille, Institut Paoli Calmette), from January 2011 to December 2018, who underwent Pancreaticoduodenectomy (PD) for resectable PDAC. Preoperative CT scan was used to measure sarcopenia and visceral fat according to international cut-offs. Neutrophil to lymphocyte (NLR) and platelet to lymphocyte ratios (PLR) were used to measure inflammation. For univariate and multivariate analyses, the Cox proportional-hazard model was used. P-values below 0.05 were considered significant.

Results

Sarcopenic patients with visceral obesity were less likely to survive than the others in multivariate analysis (OS, HR 1.65, p= 0.043). Cutaneous obesity did not influence survival. We also observed an influence on survival when we studied sarcopenia with visceral obesity (OS, p= 0.056; PFS, p = 0.014), sarcopenia with cutaneous obesity (PFS, p= 0.005) and sarcopenia with PLR (PFS, p= 0.043). This poor prognosis was also found in sarcopenic obese patients with high PLR (OS, p= 0.05; PFS, p= 0.01).

Conclusion

Sarcopenic obesity was associated with poor prognosis after PD for PDAC, especially in patients with systemic inflammation. Pre operative management of these factors should be addressed in pancreatic cancer patients.

Introduction

Pancreatic adenocarcinoma (PDAC) is the third leading cause of cancer death, with a five-year survival rate of 12 % [1]. Despite advances in chemotherapy and radiotherapy, it is projected to become the second leading cause of cancer- related mortality by 2030 [2]. The underlying cause of PDAC remains unknown, but several risk factors such as genetic predispositions, smoking, alcohol use, chronic pancreatitis, diet, and obesity have been identified [3, 4]. To this day, surgical resection accompanied with perioperative chemotherapy remains the only curative option to treat PDAC [5]. However, this approach is linked to high morbidity and mortality rates [6, 7].

When it comes to surgical treatment of various cancer types, anthropometric factors have been shown to influence patient postoperative survival [8–10]. Notably, some studies have shown that obesity and sarcopenia in PDAC are associated with lower survival rates [11, 12]. According to the European working group on sarcopenia in older people (EWGSOP), sarcopenia is defined by low muscle mass and low muscle strength and is assessed using different cut-offs [13]. Obesity, defined by a body mass index (BMI) superior to 30 kg/m² is associated with poor health outcomes such as cancer [14]. Sarcopenic obesity is prevalent in today’s population and appears to be increasing with age [12, 15] and measurable with CT-Scan [16].

There has been a recent focus on sarcopenia and inflammation with regards to several malignancies. Previous reports have shown that these factors are interrelated and that they are predictors of poor patient outcomes [17, 18]. Recently, platelet to lymphocyte ratio (PLR) and neutrophil to lymphocyte ratio (NLR) markers have been effectively used to identify systemic inflammatory response (SIR) and to predict post-operative survival in PDAC patients [19].

The aim of this study is to demonstrate the impact of sarcopenia, visceral obesity, sarcopenic obesity, and inflammation status on postoperative oncological outcomes after pancreatoduodenectomy (PD) for PDAC.

Patients and methods

Study population

Participants were identified from the department of digestive and transplant surgery at Lille University Hospital and from the department of surgical oncology at Institut Paoli-Calmette in Marseille. We selected patients who underwent PD for PDAC from the 1^st of January 2011 to the 31^st of December 2018. Only patients with PDAC comprising of surgically resectable tumors were included in our study. Underage patients, patients treated with neoadjuvant chemotherapy and/or radiotherapy, patients who underwent palliative procedures, and those with low quality CT scans were excluded from our analysis. We believe that being so selective about our patients would be more accurate to study the correlation between sarcopenia, visceral obesity and prognosis. The study complied with French National Health guidelines on research involving human subjects. IRB approval was not needed for this study.

Data collection

Clinical data including demographics, ASA score, Charlson comorbidity score, preoperative body weight, height, BMI, laboratory data, tumor characteristics were retrieved from patients’ medical records. Recorded data comprised of: intraoperative (operative time, blood loss, type of pancreatic anastomosis, vascular reconstruction), histological (histologic classification, degree of differentiation, and resection margin status), postoperative (length of stay, post-operative complications as defined by the Clavien-Dindo classification, and pancreatic surgery complications as defined by guidelines of International Study Group of pancreatic surgery) [20], and follow-up data (adjuvant therapy rate, recurrence rate, overall survival, and disease-free survival).

Muscle mass, subcutaneous and visceral adipose tissue measurements

Preoperative CT scans were performed at least one month prior to the surgery and analyzed using the MYRIAN software. Total abdominal muscle area (TAMA) comprising of transversus abdominis, external and internal obliques, rectus abdominis, psoas muscles and para spinal muscles, was measured on one image slice at vertebral level L3. Subcutaneous fat area (SFA) and visceral fat area (VFA) were also measured using this method. We calculated skeletal muscle index (SMI) as follow: TAMA measured at L3 (cm²) divided by height (m²). Since there is no gold standard cut off for sarcopenia, we used Prado’s cut-offs, based on BMI and SMI. Indeed, their study held a large variability of body composition in cancer patients with statistics to support their number choices [13] : obese (BMI>30 kg/m²) women with SMI<38.5 cm²/m², non-obese (BMI<30 kg/m²) women with SMI<32 cm²/m², obese men with SMI<52.4 cm²/m², and non-obese men with SMI<42 cm²/m². Visceral obesity was defined as visceral fat area VFA>130 cm² measured at L3 level. Subcutaneous obesity was classified as high for patients with subcutaneous fat area SFA>177 cm². In line with other studies, median measures were used as gold-standard cut-offs have yet to be defined [21, 22].

Inflammatory markers

Biological data were collected retrospectively from six-month old laboratory data. As reported in previous studies, NLR>3 was defined as high, and the PLR cut-off was 150 [23–26].

Statistical analysis

Computation was performed using the R software, with significant P-values being lower than 0.05. Continuous variables were compared using student t-test and categorical variables were analyzed using Fisher’s test. The Kaplan-Meier method was used to calculate survival curves that were then compared with the log-rank test. Univariate and multivariate analyses were performed using the Cox proportional-hazard model.

Results

A total of 364 patients underwent PD during the study period. Among them, 88 patients were excluded due to unavailable preoperative CT scans. Of the 276 remaining patients, 20 were excluded as they were previously treated with neoadjuvant therapy and 24 were excluded as their treatment comprised of a procedure different from PD. Of the 232 patients included in our study, 49% were sarcopenic and 10% suffered from sarcopenic obesity. Sarcopenic patients were more likely to have low BMI, SMI, and cutaneous adipose tissue (Table 1).

Table 1.

Characteristics of the study population

Caracteristics	All patients n = 196			Sarcopenia
	N			No ; n = 98		Yes ; n = 94
Age$	196	67,47	9,82	67,73	8,69	67,63 (+/10,36)	10,36
Sex (n male %)**	195	108	55,4	37	37,8	68	72,3
ASA Score$	196	2,11	0,64	2,08	0,65	2,15	0,62
Score 0-2 (n(%))		149	76	76	77,6	69	73,4
Score 3-4 (n(%))		47	24	22	22,4	25	26,6
Charlson$	115	2,96	1,55	3,24	1,49	2,73	1,59
Score 0-2 (n(%))		45	39,13	18	30,5	24	46,15
Score 3-4 (n(%))		52	45,22	28	47,5	23	44,23
Score >4 (n(%))		18	15,65	13	22	5	9,62
Medical history (n(%))
High blood pressure	133	63	47,4	37	54,4	23	37,7
Ischemic heart disease	114	12	10,5	8	13,8	4	7,7
Heart failure	114	1	0,9	0	0	1	1,9
Kidney failure	114	1	0,9	1	1,7	0	0
COPD*	115	4	3,5	4	6,8	0	0
Respiratory failure	114	0	0	0	0	0	0
Chronic pancreatitis	113	6	5,3	3	5,3	3	5,8
Diabetes	133	31	23,3	18	26,5	12	19,7
Cirrhosis	103	1	1	1	2	0	0
Discovery circumstances (n(%))
Abdominal pain	112	57	50,9	31	54,4	25	49
Jaundice	114	80	70,2	41	70,7	36	69,2
Acute cholitis	110	8	7,3	3	5,5	4	7,8
Worsening diabetes	111	10	9	7	12,5	3	5,9
Pancreatitis	113	19	17,1	8	14,3	11	21,6
Malnutrition (n(%))	115
moderate		51	44,35	23	39	26	50
severe		29	25,22	15	25,4	13	25
BMI$**	192	24,54	4,13	26,10	4,34	22,82	3,19
SMI$**__	193	46,46	13,08	52,77	12,48	39,87	10,23
Sarcopenia (n (%))	192	94	49	0	0	1	100
Cutaneous fat (cm2/m2)$**	196	197	99	240,46	102,96	148,73	68,36
Visceral fat (cm2/m2)$*	193	146	94	159,67	95,22	129,45	88,03
Ca 19-9$	81	495,79	851	430,37	874,58	562,04	853,97
Albumin$	115	37,03	6,24	37,7	5,86	36,81	6
Pre-albumin$	114	0,19	0,07	0,19	0,07	0,20	0,08
CRP$	85	17,90	29,90	16,77	25,6	19,10	34,26
NLR$	161	3,50	2,35	3,17	2,26	3,78	2,40
PLR$	160	181,09	98,22	171,4	84,85	191,3	109,08
Post operative complications :
Clavien Dindo > 3a (n(%))	130	28	21,5	17	24,6	11	18,6
Anatomopathology
pT (n(%))	192
pT1-pT2		51	26,6	30	31,2	21	22,6
pT3-pT4		141	73,4	66	68,8	72	77,4
Vascular embolus (n(%))	154	65	42,2	28	36,4	36	48,6
Lymphatic embolus (n(%))	108	20	18,5	9	16,7	9	17,6
Resection (n(%))	194
R0		169	87,1	86	87,8	80	86
R1		24	12,4	11	11,2	13	14
R2		1	0,5	1	1	0	0
N+$	189	2,47	2,64	2,37	2,75	2,54	2,5
N+/N$	189	0,19	0,43	0,20	0,58	0,17	0,18
Adjuvant chemotherapy (n(%))	112		83,9	46	80,7	44	86,3
Mortality at 90 days (n(%))	196	8	4,1	5	5,1	3	3,2
Dead (n(%))	196	135	69	64	65,3	69	73,4
Days before death$	196	636	478	656	534	626	427
Survival in days$	196	1074	876	1164	959	969	764

ASA American society of anesthesiologists, COPD Chronic obstructive pulmonary disease, BMI Body mass index, SMI Squeletal muscle index, NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio

^$= mean (+/- SD)

^*= p-value < 0,005

^**= p-value < 0,001

We studied the correlation between sarcopenia and inflammatory markers. No correlation was observed between sarcopenia and NLR or PLR. However, sarcopenic patients tended to present with higher NLR and PLR compared to non-sarcopenic patients.

Moreover, we assessed the impact of sarcopenia, NLR and PLR on oncological outcomes after PD for PDAC. Sarcopenic patients were shown to have a lower progression-free survival (PFS) compared to non-sarcopenic patients (p=0.09). However, overall survival (OS) was similar for sarcopenic patients and non-sarcopenic patients (p=0.2). Regarding inflammatory status, no significant differences were observed in PFS and OS between patients with high NLR and those with low NLR. Similar results were observed regarding PLR status. With regards to obesity status, OS and OFS were not impacted by VFA and SFA after PD for PDAC.

We then investigated patients’ survival outcomes based on their sarcopenia and inflammatory status. We found that sarcopenic patients with high NLR tended to have worse prognosis than non-sarcopenic patients with high NLR, but that the difference was not significant (Fig. 1A). However, PFS was significantly lower for sarcopenic patients with high PLR compared to non-sarcopenic patients with low PLR (p=0.043). Moreover, OS appeared to be lower for sarcopenic patients with high PLR compared to non-sarcopenic patients with low PLR (p=0.096) (Fig. 1B).

Fig. 1.

Survival outcome based on sarcopenia and inflammatory status, A Survival in sarcopenic and non-sarcopenic patients with high or low NLR, B Survival in sarcopenic and non- sarcopenic patients with high or low PLR

Furthermore, we quantified the impact of sarcopenia and obesity on survival. Patients who were classified as sarcopenic and viscerally obese demonstrated lower survival rates compared to non-sarcopenic and non-obese patients (OS, p= 0.014; PFS, p=0.05) (Fig. 2A). Furthermore, sarcopenic patients with cutaneous obesity had a lower PFS (p = 0.0059) compared to non-sarcopenic and non-cutaneous obese patients (Fig. 2B).

Fig. 2.

Survival outcome based on sarcopenia and obesity, A Survival in sarcopenic and non-sarcopenic patients with visceral obesity, B Survival in sarcopenia and non-sarcopenic patients with cutaneous obesity

Among our study participants, 20 were considered sarcopenic, viscerally obese, and had a high NLR. Although the difference is not significant, these patients had lower rates of OS and PFS compared to non-sarcopenic, non-obese patients with low NLR. Moreover, sarcopenic obesity in patients with high PLR had a lower PFS (p=0.01) and OS (p=0.05) compared to non-sarcopenic, non-obese patients with low PLR (Fig. 3).

Fig. 3.

Survival outcome based on sarcopenia, visceral obesity and PLR status, A Progression free survival, B Overall survival

Finally, a multivariate analysis with Cox proportional-hazard ratio was performed to identify the risk factors of worse prognosis after PD for PDAC. Based on the multivariate analysis, the sarcopenic visceral obesity variable was independent from poor prognostic factors. The result was significant for OS (HR, 2.17; 95% CI,1.2–4.6, p=0.04) (Fig. 4A) and PFS (HR, 2.34; 95% CI,1.16–4.7, p=0.018). Lastly, lower PFS was associated with high PLR and cutaneous obesity (p= 0.08 and p= 0,064 respectively) (Fig. 4B).

Fig. 4.

Multivariate analysis (COX Proportional Hazard ratio model, p= 0,05), A Overall survival, B Progression free Survival

Discussion

This study is the first to examine the relationship between SIR, sarcopenia, and visceral obesity in PDAC. Our results demonstrate an interdependence between inflammation and sarcopenia and highlight their impact on positive patient outcomes. Interestingly, we found that sarcopenic patients with visceral obesity were more likely to relapse following PDAC and presented reduced post-operative survival. Our data suggests that preoperative evaluation of sarcopenia, inflammation and visceral obesity status could help to determine postoperative patient outcomes and thus assess if a patient is an appropriate candidate for surgery.

Based on our results, sarcopenia alone does not impact oncological outcomes after PD for PDAC (OS, p= 0.2; PFS, P= 0.09). Several papers have identified sarcopenia as a risk factor for poor prognosis in many cancers including PDAC [10]. Previous research suggest that tumor aggressiveness is correlated with sarcopenic status [26]. Leads on decreasing sarcopenia have been established, but the data is lacking for patients undergoing PD. Physical training against resistance and improved nutrition (nutritional supplements by vitamin D, ß-methylbutyrate and creatine) up to three months before surgery have been proposed to enhance post-operative survival [27, 28]. Several randomized controlled studies to identify an adequate preoperative program for sarcopenic patients before highly complex procedures are in progress.

In cancer patients, sarcopenia is frequently associated with inflammation. Indeed, tumors create an inflammatory environment by producing diverse cytokines, altering local balance, and attracting neutrophils [29]. Moreover, chronic inflammation generates neoplastic processes by inducing DNA damage [30]. It has been suggested that PLR is a useful prognostic factor in some cancer types since platelets enhance tumor development by promoting angiogenesis through the cytokine vascular endothelial growth factor (VEGF). Due to the involvement of lymphocytes and neutrophils in PDAC and the tumoral microenvironment, NLR has also been associated with PDAC survival [31, 32]. NLR has been used as a prognostic factor in several cancers [33–35] including PDAC [35]. When NLR is highly elevated, surgery should be delayed, and neoadjuvant treatment should be prioritized. In fact, neoadjuvant chemotherapy has been shown to lower NLR and improve postoperative prognosis in patients with breast cancer [34]. Moreover, the normalization of PLR and NLR after the first cure of FOLFOX in patients with gastric cancer was associated with better OS [36]. As previously reported, these findings indicate that sarcopenia and inflammation reflect aggressive tumor process [16, 37, 38].

Our investigation also focused on the impact of VFA and SFA on PDAC prognosis. Our results suggest that visceral and subcutaneous obesity do not impact patient survival. Nonetheless, sarcopenic and viscerally obese patients had poorer OS and PFS, regardless of their BMI. Similarly, Okumura et al. reported that sarcopenic visceral obesity is closely associated with mortality and recurrence [16, 39, 40]^. These findings demonstrate that the redistribution of muscle mass and adipose tissue is a key factor to consider when predicting patient prognosis. An emerging topic about fat distribution is myosteatosis and its association with sarcopenia. More studies are reporting the influence of muscle fat on muscle strength and physical activity leading to a worse post operative prognosis in cancer patients regardless of sarcopenia [41, 42].

Our results highlight the prognostic role of sarcopenic visceral obesity associated with inflammation in patients with PDAC. Visceral fat increases proinflammatory cytokine levels, leading to chronic inflammation and consequently increased carcinogenesis [43, 44]. Moreover, adipokines and myokines affect the immune system and promote inflammation [43]. Thus, visceral adiposity, loss of muscle mass, and inflammation are interrelated factors that predict poor patient outcomes [45].

Our study presented some limitations. First, the retrospective study design can be subject to bias, suggesting the need for a prospective study in the future. Second, cut-offs to define sarcopenia, visceral obesity, and inflammation in patients with PDAC were not previously validated due to the lack of a current consensual definition [46]. Third, our analysis only focused on patients undergoing PD, preventing generalization of our results to other treatments or surgical interventions. Finally, sarcopenia was assessed based on preoperative CT scan by measuring the TAMA. In the future, more effective methods could be used to determine sarcopenic state such as dual X-ray absorptiometry (DXA) or bioelectrical impedance analysis (BIA) [47].

Sarcopenic obesity and SIR are associated with poorer patient outcomes and lower survival rates. Our results suggest that the prognosis of pancreatic cancer patients can be improved with various treatment adjustments. Sarcopenia, visceral obesity, and inflammatory status being predictive of poor prognosis, patients who are assessed for these factors should be considered borderline resectable PDAC and should be treated with neoadjuvant chemotherapy [45]. When sarcopenia is screened for in daily practice, the importance of nutritional care and exercising should be explained to patients. Nutrition can be improved with adequate protein intake of 1.2 g/kg/day. Progressive resistance training (PRT) [48] and supplementation can improve muscle mass, muscle strength and physical performance [47, 49, 50]. Our analysis provides a framework for assessing patient preoperative parameters to determine the most appropriate treatment approach for resectable PDAC. Several protocols can be implemented to improve patient survival and should include exercise,

nutrition modification or neoadjuvant chemotherapy. Further studies should be performed to assess the efficiency of these measures in determining PDAC treatment outcomes.

Authors’ contributions

KB wrote the entire Manuscript. MEA contributed by helping with the structure of the research, the data, the in-site reviewing. YR contributed by reading and correcting the final form of the manuscript. CAB contributed by helping doing the statistics after the data were collected. ST, GP, OT and JG contributed by letting me access the patients' data.

Funding

No fundings were received by authors for this word.

Availability of data and materials

Some data are available.

Data availability

Data that support the findings of this study were collected from the onsite hospital software (in Lille and Marseille).

Declarations

Ethics approval and consent to participate

The study complied with French National Health guidelines on research involving human subjects. IRB approval was not needed for this study.

Competing interests

The authors declare no competing interests.