Pancreatic cancer remains one of the most lethal diseases, with a tendency to reach in the next decade one of the leading causes of cancer mortality in Western countries. The five-year survival is approximately 8%, with more than half of patients being diagnosed in stage IV. Until the introduction of gemcitabine in 1997, 5-fluorouracil was the therapeutic standard, the median survival being about 4.4 months, increasing by about 1.3 months after gemcitabine was introduced as a therapeutic standard alone.
FOLFIRINOX – past, present and future perspectives
An association of chemotherapies (5-fluorouracil, leucovorin, irinotecan and oxaliplatin), the FOLFIRINOX regimen demonstrated in a phase II trial a survival benefit, reaching 10.2 months, with an improvement in the quality of life, but with an increased hematological toxicity (grade 3 and 4 neutropenia) in more than 50% of cases. The toxicity profile appears to differ significantly depending on the treated population. The study by Okusaka et al. reported, on a Japanese population, a rate higher than 75% for grade 3 and 4 neutropenia, and almost 89% of the patients needed to reduce or delay the dose administration of FOLFIRINOX. This study led to the first-line approach of FOLFIRINOX in Japan in 2013 for unresectable pancreatic cancer. Two years before, the PRODIGE 4/ACCORD 11 trial, which enrolled 342 patients from France with Eastern Cooperative Oncology Group (ECOG) performance status 0 or 1, was the basis for the approval of the FOLFIRINOX regimen for patients with unresectable pancreatic cancer with good performance status(1).
A study that analyzed the tolerance and efficacy of the dose-adjusted FOLFIRINOX regimen in elderly patients diagnosed with colorectal or pancreatic cancer demonstrated the feasibility and safety of this regimen, if the adjustment of irinotecan (dose according to the UGT1A1 polymorphism) is associated with 5-fluorouracil (dose according to the dihydropyrimidine dehydrogenase pharmacodynamics). The correlation between irinotecan toxicity and heterozygous UGT1A1 polymorphisms was also demonstrated by Shirasu et al. In patients treated with the initial FOLFIRINOX regimen, the level of toxicity is lower than in patients with wild-type UGT1A1. A study that evaluated the possibility of using the FOLFIRINOX regimen concluded that, for patients over 70 years of age, a modified FOLFIRINOX regimen could be used due to the absence of bolus and to irinotecan dose reduction, the multivariate analysis demonstrating that the geriatric risk factors do not affect the survival(2,3).
In the adjuvant treatment, fluorouracil-based chemotherapy has become the standard treatment, following the publication of data from a 15-year trial that revealed a survival benefit for patients receiving adjuvant fluorouracil chemotherapy(4).
The trial that modified the FOLFIRINOX regimen (oxaliplatin 85 mg per square meter of body-surface area, irinotecan 150 mg per square meter, leucovorin 400 mg per square meter, and fluorouracil 2400 mg per square meter), every two weeks, was compared with the baseline regimen based on gemcitabine 1000 mg/square meter on days 1, 8 and 15, every four weeks, for 24 weeks(5).
Surgery remains the basic treatment of unresectable pancreatic cancer. Even though an R0 resection is possible, the recurrence rate is close to 80%. Numerous studies have tried to establish the optimal therapeutic behavior: radiotherapy, chemotherapy or the combination have been evaluated in adjuvant settings. The first study conducted by the North American Gastrointestinal Tumor Study Group has shown the superiority of chemotherapy. The European Study Group of Pancreatic Cancer (ESPAC) has finally shown a benefit in survival for adjuvant chemotherapy but not for chemotherapy, the patients who received 20 Gy and 5FU presenting a detriment in survival. Subsequently, the patients receiving protocols based on mytomycin C and doxorubicin demonstrated insignificant survival benefits. The ESPAC-3 trial, that included 1088 patients, demonstrated an advantage for chemotherapy (23.6 months versus 23 months) regarding survival for gemcitabine, but an improvement in the safety profile compared to 5FU/folinic acid, confirming the gemcitabine-based chemotherapy standard treatment. These results confirm the data obtained in the CONKO-001 trial which demonstrated a net benefit for gemcitabine versus observation regarding DFS and OS (20.2 months versus 22.8 months). The ESPAC-4 trial demonstrated a benefit of 2.5 months for the combination of oral fluoropyrimidines, but also increased toxicity. The maximum benefit was found for R1 resections, with OS rates of 54.4 months in the group receiving mFOLFIRINOX in favor of the gemcitabine-treated group for which OS was 35 months. In the FOLFIRINOX-treated group, the grade III and IV adverse effects were 75.9%, while in the gemcitabine-treated group, only 29% were reported. Thus, the mFOLFIRINOX protocol becomes a therapeutic standard in the adjuvant treatment of pancreatic cancer(6-8).
In the case of pancreatic cancer, the authors have analyzed the possibility of de-escalating the FOLFIRINOX protocol to FOLFOX regimen, for cases with low treatment tolerance or non-progression disease. The maintenance with 5FU/leucovorin was used after 6 cycles and FOLFIRINOX will be restarted at the time of a disease progression. The authors consider the method feasible in reducing toxicities by ensuring a prolonged progression-free survival (PFS)(9).
In 2011, Conroy et al. demonstrated that a triple combination regimen, including leucovorin, oxaliplatin, irinotecan and fluorouracil, improved both overall survival and the quality of life, compared to the gemcitabine-based regimen, with an OS of 11.1 months versus 6.8 months for patients treated with gemcitabine. An improvement of life quality was also noticed; however, this regimen was associated with increased toxicities (neutropenia, diarrhea, peripheral neuropathy). The association of gemcitabine with nab-paclitaxel was also approved in the first-line treatment of pancreatic cancer, even though the OS and PFS data are lower, with 2.6 months for OS and 3.3 months for PFS, than those for the FOLFIRINOX regimen(1,10).
The nab-paclitaxel-gemcitabine regimen has an advantage in terms of toxicity profile compared to the FOLFIRINOX regimen proposed by Conroy in 2011. The high toxicity of the FOLFIRINOX profile made it necessary to evaluate the therapeutic protocols with an improved toxicity profile, but with efficacy and tolerance similar to the initial regimen, as Cavanna et al. reported using a chemotherapy scheme with -20% bolus 5-FU and -25% irinotecan(11,12).
Even though the combination therapy with oxaliplatin, irinotecan, fluorouracil and leucovorine (FOLFIRINOX) has been shown to bring a significant survival benefit for patients with locally advanced or metastatic pancreatic cancer, the efficacy of the FOLFIRINOX regimen after the failure of a gemcitabine treatment line was poorly investigated. A phase II study with a single prospective arm evaluated the feasibility and safety of the attenuated FOLFIRINOX regimen in this situation. The study included patients with ECOG performance status 0 and 1. The protocol included oxaliplatin 65 mg/m2, irinotecan 135 mg/m2 and leucovorine 400 mg/m2 injected intravenously on day 1 and 5-fluorouracil 2000 mg/m2 continuously intravenously infused over 46 hours on days 1 and 2, repeated after two weeks. The disease control was obtained in 64.1% of cases, the rates of OS at six months and one year were 59% and 15.4% respectively, and OS was 8.5 months (between 5.6 and 11.4 months). The most common grade 3 or 4 toxicity was neutropenia (41%), and the authors concluded that attenuated FOLFIRINOX regimen is a feasible alternative for patients refractory to gemcitabine treatment(13).
The identification of predictive biomarkers of response to FOLFIRINOX treatment has become a topical issue, with the increasing use of this toxic regimen in the neoadjuvant treatment of borderline resectable cancers. Although there are four studies that attempted to identify biomarkers of treatment response, no biomarker has been identified and validated up to the present. Numerous studies including liquid biopsy are investigating this topic which is becoming more and more important in the context of the need to identify those groups of patients who benefit from an aggressive regim(14).
The FOLFIRINOX protocol is also evaluated in a multicenter phase III trial (IDEA) that includes patients with R0 resected colon cancer with curative intent of a pT4N1 or pT1-4N2 colon adenocarcinoma. The study randomized patients into two arms receiving 12 cycles of modified FOLFOX6 (oxaliplatin 85 mg/m2, leucovorin 400 mg/m2, 5-FU bolus 400 mg/m2, then 2.4 g/m2 over 46 hours) at two weeks during 24 weeks or modified FOLFOX (oxaliplatin 85 mg/m2, leucovorine 400 mg/m2, irinotecan 180 mg/m2 and 5-FU 2.4 g/m2 over 46 hours) as adjuvant treatment. The results of the study will be available after 5 years of follow-up, expecting an advantage of about 10% in DFS, but also a benefit in OS(15,16).
The concept of induction and re-induction was used in the OPTIMOX 1 trial on colorectal cancers, on one of the arms using maintenance with 5FU/leucovorin after six cycles of FOLFOX protocol chemotherapy and oxaliplatin being reintroduced to progression. The European Study Group of Pancreatic Cancer (ESPAC) has shown a benefit in survival for adjuvant chemotherapy but not for radiochemotherapy. The patients who have received radiotherapy in a total dose of 20 Gy and 5FU-based chemotherapy presented a detriment in survival. Subsequently, the patients’ protocols based on mitomycin C and doxorubicin demonstrated insignificant survival benefits(17).
Perri and colleagues comparatively analyzed the therapeutic response and survival of pancreatic cancer patients for the first-line chemotherapy based on gemcitabine, the FOLFIRINOX protocol or nab-paclitaxel. Evaluating the objective response according to RECIST imaging criteria, serum level variation of CA 19-9 antigen and OS, the authors did not identify major differences in OS if FOLFIRINOX or nab-paclitaxel was administered. The response rate according to the RECIST criteria and the rate of pancreatectomies were higher in the group treated with the FOLFIRINOX regimen. However, there were differences in the heterogeneity of the study group, that included 485 patients, the FOLFIRINOX regimen being more frequently administered in patients with a more favorable ECOG performance status, and nab-paclitaxel being administered more frequently in patients with more invasive tumors. A controversial topic remains the value of adjuvant chemotherapy up to three months after pancreatectomy in patients who have received neoadjuvant chemotherapy with the FOLFIRINOX protocol. Analyzing a group of 520 patients, of which 66% received adjuvant chemotherapy in the FOLFIRINOX protocol and the rest chemotherapy based on gemcitabine, capecitabine, therapeutic combinations or other agents, Roessel et al. identified an adjuvant benefit for OS only if there was evidence for pathological node positive disease, but the rate of neoadjuvant response, R0 resections and the duration of each sequence were not taken into account and should be evaluated in the future clinical trials. However, the systematic review and network meta-analysis of Parmar demonstrates a net benefit for mFOLFIRINOX in adjuvant OS, with no evidence of a benefit of chemotherapy based on gemcitabine, S1, nab-paclitaxel or if a combination is used, including erlotinib. New strategies have also been tested in addressing metastatic pancreatic cancer. The combination of chemotherapy with irinotecan, oxaliplatin and S-1, an oral prodrug of 5FU, has demonstrated safety and efficacy in terms of toxicity and OS profile(18-21).
The stratification into molecular subgroups, with therapeutic consequences for pancreatic cancer cases, led to the idea of analyzing the response to different FOLFIRINOX, depending on the molecular subtypes, the most relevant being considered, in terms of therapeutic implication, the KRT81 quasi-mesenchymal (QM)/squamous/basal-like and HNF1A non-QM, overlap with exocrine/ADEX subtype. In the group of patients in which the response to chemotherapy was evaluated, the authors of the study noted a lack of survival difference between the double-negative subtype and the HNF1A-positive subtype, the positive KRT81 being associated with a poor prognosis of the patient. KRT81-positive did not benefit from FOLFIRINOX therapy, and in HNF1A-positive, FOLFIRINOX group was superior in response to gemcitabine(22).
Recently, the results of the POLO trial revealed a significant benefit in PFS for the use of PARP inhibitors in maintenance after at least 16 weeks without progression for patients receiving platinum-based chemotherapy. The OS benefit remains modest. The trial included only patients with metastatic pancreatic cancer with the BRACA 1 and BRCA 2 mutations, present in less than 10% of the patients treated for pancreatic cancer(23).
Starting from the idea that most cases of pancreatic cases overexpress EGFR pathway, the use of molecules that block this pathway has been investigated in clinical trials. The results of using erlotinib and cetuximab were controversial or had minor benefits for a subgroup in the study population(24).
Immunotherapy is currently being evaluated in clinical trials as a single treatment or in combination. The results obtained so far are disappointing. PD-L1 and CTLA4 inhibitors demonstrated inefficiency alone. Among the theories that explain this, the most feasible is related to poor tumor immunogenicity, due to the small number of mutations (approximately 45), but also to the tumoral immunosuppressive microenvironment containing cancer-associated fibroblasts. The validation of some biomarkers of the response to immunotherapy and the identification of the possibility of increasing the immunogenicity of the cancerous tumor cell will be the basis of a future clinical application of immunotherapy for pancreatic cancer(25).
Erkan et al. identifies tumor hypothesis as a factor associated with invasive pancreatic cancer phenotype, also being correlated with resistance to adjuvant chemotherapy treatment. Hypoxia was also correlated with early metastasis. Thus, the therapeutic strategies that focus on reducing tumor hypoxia are part of the future directions in the treatment of pancreatic cancer(26).
Eight years after its introduction into clinical practice, the FOLFIRINOX protocol remains the basis of the first-line treatment of locally advanced and metastatic pancreatic cancer, proving its efficacy in the non-adjuvant and adjuvant setting in this pathology. Although the therapeutic spectrum of pancreatic cancer will be improved by identifying biomarkers and new therapeutic targets, FOLFIRINOX remains the most efficient treatment for pancreatic cancer patients with ECOG performance status of 0 or 1, proving, from the perspective of new data from the clinical studies, a therapeutic potential in colorectal cancer. In the context of the disappointing results of targeted therapy and immunotherapy alone, future strategies will focus on associating these therapeutic methods with chemotherapy, also identifying biomarkers to stratify the treatment. The use of FOLFIRINOX in modified regimens or an attenuated toxicity in sequential association with immunotherapy, EGFR and PARP inhibitors, and the use in clinical trials of other therapeutic targets are options that may be explored in the future. FOLFIRINOX regimen remains the cornerstone of systemic treatment for one of the deadliest types of cancer.
Conflicts of interests: The authors declare no conflict of interests.