Evaluation of predictive factors of toxicity of chemotherapy with FOLFIRINOX in patients treated for pancreatic adenocarcinoma

Pancreatic cancer is the 7th leading cause of cancer death worldwide and its incidence is on the rise. Chemotherapy with FOLFIRINOX is one of the metastatic first-line reference regimens. This treatment is sometimes a source of severe toxicity that may require hospitalization and subsequent dose reductions. The objective of this work is to evaluate the biological, clinical and anthropometric factors that can predict the occurrence of severe toxicity in FOLFIRINOX.

This is a retrospective study let at Gustave Roussy Cancer Campus in France and conducted in all patients with pancreatic adenocarcinoma treated with FOLFIRINOX between 2010 and 2021. Biological, clinical and anthropometric characteristics were collected before the start of the FOLFIRINOX and hospitalizations/tolerance (graded according to the CTCAE v5.0 classification) collected during the first month of chemotherapy. Anthropometric data were all evaluated from pre-treatment scans and with the help of the Anthropometer3DNet software. Our study deliberately focused on the early onset toxicity.

In our study, 152 patients with a median age of 62 years, suffering from pancreatic cancer and treated with FOLFIRINOX were included. FOLFIRINOX was administered in the metastatic, locally advanced or adjuvant/neoadjuvant setting in 123 patients (81%), 22 patients (14%) and 7 patients (5%) respectively. The patients had a general condition assessed by Performance Status (PS) 0, 1, ≥2 in 74 patients (49%), 62 patients (41%), 15 patients (10%) and unspecified in 1 patient. In our series, the median follow-up was 62.5 months, the median overall survival (OS) 13.7 months and the median progression-free survival (PFS) 8.9 months. Dose reduction prior to the first cycle was performed in 21 patients (14%). During the 1st month of treatment (i.e. after a maximum of 2 cycles), 73 patients (48%) experienced toxicity resulting in hospitalization and/or dose reduction at subsequent courses and 43 patients (28%) required hospitalization for a median of 8 days. Among the anthropometric variables studied, the presence of low muscle mass (MBM) was significantly correlated with the need to reduce chemotherapy doses (AUC 0.63 [95% CI: 0.55 - 0.75]; p < 0.001). Other variables such as Subcutaneous Adipose Tissue (SAT), Visceral Adipose Tissue (VAT), Lean Body Mass (LBM), Fat Body Mass (FBM) were not predictive of toxicity. Among the biological variables, only the neutrophil/lymphocyte ratio (NLR) greater than 4 was significantly correlated with the need to reduce chemotherapy doses and with hospitalization. Among all variables studied, only an NLR ratio of less than 4 was significantly correlated with longer OS (p=0.003).

In our retrospective cohort of pancreatic cancers treated with FOLFIRINOX, toxicities requiring dose reduction and/or hospitalization were frequent (48%) and their occurrence correlated with the presence of low muscle mass and/or high NLR. This study requires confirmation in a larger series but suggests that the assessment of muscle mass and NLR ratio could limit the toxicities of FOLFIRINOX chemotherapy.

The authors.

Has not received any funding.

A. Hollebecque: Honoraria (Institution): Astra-Zeneca, Roche; Advisory / Consultancy: Abbvie, Amgen, Basilea, BMS, Incyte, Servier, QED therapeutics, Relay Therapeutics, Seagen, Taiho; Research grant / Funding (self): Incyte; Research grant / Funding (institution): AstraZeneca; Travel / Accommodation / Expenses: AstraZeneca. All other authors have declared no conflicts of interest.