FOLFIRINOX as induction treatment in rectal cancer patients with synchronous metastases: Results of the FFCD 1102 phase II trial

European Journal of Cancer 104 (2018) 108e116

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.ejcancer.com

Original Research

FOLFIRINOX as induction treatment in rectal cancer patients with synchronous metastases: Results of the FFCD 1102 phase II trial*,** J.B. Bachet a,*, O. Lucidarme b, C.B. Levache c, E. Barbier d, J.L. Raoul e, T. Lecomte f, C. Desauw g, F. Brocard h, S. Pernot i, G. Breysacher j, J.P. Lagasse k, F. Di Fiore l, P.L. Etienne m, O.J.M. Dupuis n, A. Aleba o, C. Lepage p, J. Taieb i, for FFCD 1102 investigators1 a

Sorbonne Universite´, UPMC Universite´, Department of Hepato-Gastroenterology and Digestive Oncology, Groupe Hospitalier Pitie´ Salpeˆtrie`re, APHP, Paris 06, Paris, France b Sorbonne Universite´, UPMC Universite´, Department of Radiology, Groupe Hospitalier Pitie´ Salpeˆtrie`re, APHP, Paris 06, Paris, France c Department of Oncology, Pe´rigueux, France d Biostatistics, FFCD, EPICAD INSERM LNC-UMR 1231, University of Burgundy and Franche Comte´, Dijon, France e Department of Oncologie, Centre Anti-cance´reux Paoli-Calmettes, Marseille, France f Department of Hepato-Gastroenterology, Centre Hospitalier Re´gional Universitaire de Tours, Hoˆpital Trousseau, Tours, France g Department of Oncology, Centre Hospitalier Re´gional Universitaire de Lille, Hoˆpital Claude Huriez, Lille, France h Department of Oncology, Centre D’oncologie de Gentilly, Gentilly, France i Sorbonne Paris Cite´, Paris Descartes University, Department of Hepato-Gastroenterology and Digestive Oncology, Hoˆpital Europe´en Georges Pompidou, Paris, France j Department of Hepato-Gastroenterology, Hoˆpital Pasteur de Colmar, Colmar, France k Department of Hepato-Gastroenterology and Digestive Oncology, Centre Hospitalier La Source, Orle´ans, France l Department of Oncology, Centre Hospitalo-universitaire de Rouen, Rouen, France m Department of Oncology, CARIO, HPCA PLERIN, St Brieuc, France n Department of Oncology, Centre Hospitalier de Niort, Niort, France o Department of Oncology, Clinique Victor Hugo, Le Mans, France p HepatoGastroenterology and Digestive Oncology Department, University Hospital Dijon, University of Burgundy and Franche Comte´, FFCD, EPICAD INSERM LNC-UMR 1231, Dijon, France Received 21 June 2018; received in revised form 20 August 2018; accepted 9 September 2018

*

The FFCD Group was the only sponsor of this trial. The study results were presented in part at the 2016 ASCO Annual Meeting, Chicago, Illinois, USA (Abstract 3513), and at the 2017 ESMO Annual Meeting, Madrid, Spain (Abstract 505-P). * Corresponding author: Department of Hepato-Gastroenterology and Digestive Oncology, Groupe Hospitalier Pitie´ Salpeˆtrie`re, 47-83 Boulevard de l’hoˆpital, 75651, Paris Cedex 13, France. Fax: þ331 42 16 12 38. E-mail address: [email protected] (J.B. Bachet). 1 Please see the Appendix A for a list of the FFCD 1102 Investigators/Collaborators. **

https://doi.org/10.1016/j.ejca.2018.09.006 0959-8049/ª 2018 Elsevier Ltd. All rights reserved.

J.B. Bachet et al. / European Journal of Cancer 104 (2018) 108e116

KEYWORDS FOLFIRINOX; Rectal cancer; Synchronous metastases; Induction; Local control

109

Abstract Aim of the study: The optimal therapeutic strategy in patients with rectal cancer and synchronous unresectable metastases remains unknown. We evaluated the efficacy of FOLFIRINOX induction therapy in this setting. Patients and methods: Chemotherapy-naı¨ve patients received at least 8 cycles of FOLFIRINOX. The primary end-point was the 4-month disease control (4 m DC) rate. Tumour responses were centrally reviewed and assessed by computed tomography scan for metastases (Response Evaluation Criteria in Solid Tumours criteria) and magnetic resonance imaging for rectal tumorus. With a Simon 2-stage design and a targeted (H1) 4 m DC > 75%, 65 patients were enrolled from July 2012 to February 2015: male, 78%; median age, 61 years; performance status, 0e1, 98%; liver metastases, 92%; 2 metastatic sites, 63%. Results: Fifty-six (85%) of the 65 patients received the 8 planned FOLFIRINOX cycles. The primary objective was achieved (4 m DC rate: 94%; 95% confidence interval [CI], 86.3e97.8). Primary tumour symptoms decreased from 72% at baseline to 10% at 4 months. Response rate was 86%, and a >70% primary tumour volume decrease was seen in 63% of patients. Fortyfour patients (68%) had at least one grade 3 side-effect; no toxic deaths occurred. Median follow-up was 35.0 months (95% CI, 31.3e43.7). Median progression-free survival and overall survival were 10.9 m (95% CI, 8.8e12.9) and 33.4 m (95% CI, 22.6e38.2), respectively. Conclusion: Upfront FOLFIRINOX is feasible and allows good local and distant control. It therefore offers the opportunity to choose the best therapeutic strategy for each patient and to personalise treatment according to the local and distant efficacy results of this induction step. Trial registration: Clinicaltrials.gov, NCT01674309. ª 2018 Elsevier Ltd. All rights reserved.

1. Introduction Colorectal cancer is the second most frequent human cancer in the world, and rectal cancer accounts for about one-third of cases. Rectal cancer with synchronous metastases at diagnosis is found in about 20% of cases, and treatment strategy in these patients must take into account both the primary rectal tumour and its metastases [1]. In non-metastatic rectal cancer, pre-operative chemoradiotherapy (CRT) or radiotherapy (RT) improves local disease control, generally results in downstaging of the rectal tumour and may improve sphincter preservation for CRT, with only a marginal impact on overall survival (OS) [2,3]. However, the benefit of pre-operative CRT or RT in metastatic patients has only been evaluated to date in patients with locally advanced rectal cancer and resectable synchronous liver metastases [4,5]. Metastatic spread/progression is often presented as the main risk in patients with rectal cancer and unresectable metastases, and systemic chemotherapy is recommended with the aim of prolonging OS. However, the optimal upfront treatment remains controversial as there is also a risk of developing symptoms related to the rectal tumour, with major impairment of patients’ quality of life. Although resection of the rectal tumour or CRT has been proposed to control the primary tumour even in metastatic patients [6,7], intensive chemotherapy as induction treatment is also an

option to control all tumour sites and possibly avoid unnecessary surgery [1,8,9]. In 2007, French recommendations for clinical practice suggested beginning treatment with chemotherapy in asymptomatic patients and using a combination of radiotherapy and systemic chemotherapy in patients with a symptomatic rectal tumour [10]. Since then, modern systemic chemotherapy regimens have been reported not only to increase survival [11] but also to effectively control primary rectal tumours in nonmetastatic patients [12e14]. Triplet chemotherapy regimens are associated with the highest rates of objective tumour response regardless of RAS assessment [15,16]. The aim of the FFCD 1102 phase II trial was to assess the tolerability and efficacy of induction chemotherapy with FOLFIRINOX in patients with rectal cancer and unresectable synchronous metastases. 2. Material and methods 2.1. Study design and participants The FFCD 1102 study was a multicenter non-randomised phase II trial. The study protocol was approved by the French ethics committee ‘CCP Ile de France 8’. All patients provided informed consent before study enrolment. Main eligibility criteria were as follows: age 18 years, pathologically confirmed previously untreated

110

J.B. Bachet et al. / European Journal of Cancer 104 (2018) 108e116

rectal adenocarcinoma and unresectable synchronous metastases. Other inclusion/exclusion criteria are detailed in Supplementary Appendix B (online only). Baseline clinical and laboratory evaluations no more than 14 days before inclusion and tumour assessment (CT scan and MRI) no more than 21 days before inclusion were performed.

least one radiologic evaluation were considered evaluable. Secondary end-points included the objective response rate (ORR) of metastatic disease (RECIST criteria) and of the rectal tumour (tumour volume), local symptoms, median progression-free survival (PFS), median OS and treatment tolerability.

2.2. Procedures

2.5. Statistical analyses

Patients received induction chemotherapy with FOLFIRINOX as described previously by our group in pancreatic cancer (Supplementary Appendix B); 8 cycles were planned. Systematic granulocyte colonystimulating factor (G-CSF) could be used in primary prophylaxis or secondary prophylaxis, at the investigator’s discretion. All adverse events were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE), version 4.0. Dose reduction recommendations are detailed in Supplementary Appendix B (online only). Tumour response (Response Evaluation Criteria in Solid Tumours [RECIST] v1.1) was assessed by investigators using chest-abdominal-pelvic computed tomography (CT) scan every 8 weeks (4 cycles). Patients continued treatment until unacceptable toxicity, completion of the 8 planned cycles of FOLFIRINOX, disease progression or patient withdrawal. After the 8 planned cycles of FOLFIRINOX, the therapeutic strategy was defined in a personalised fashion according to each patient’s status and a decision taken at a local multidisciplinary meeting.

According to a Simon 2-stage design with a one-sided type I alpha error of 5% and power of 90%, 54 patients had to be included to test the following hypotheses: under the null hypothesis (H0), a 4-month DC rate of 75% was not considered interesting for further investigation, whereas, under the alternative hypothesis (H1), a 4-month DC rate of >75% was considered interesting; a 4-month DC rate of >90% was hoped for. With an expected 20% dropout/non-evaluable rate at 4 months, 65 patients were recruited. The first interim analysis was conducted on 9 April 2014, after the accrual of 30 patients (25 evaluable patients). Twenty-three patients met the efficacy criterion, and therefore, the trial was continued. The second step should have been performed on 54 evaluable patients, but, in the end, 64 patients were evaluable. The decision rules were recalculated for 64 patients: If  55 patients met the efficacy criterion, the treatment was considered effective. The primary end-point was planned for the modified intention-to-treat (mITT) populationdall evaluable patientsdand was reported with their one-sided exact 95% confidence interval (CI). Analyses of tolerability were conducted in all patients who received at least one dose of treatment. The dose intensity of each drug was calculated based on the number of cycles received by each patient. The relative dose intensity was calculated as the ratio of the dose intensity to the dose intensity indicated in the protocol (obtained as the dose specified per cycle in mg/ m2). Qualitative and continuous variables were described using the usual descriptive statistics: numbers and percentages or medians with range (minemax). Survival analyses were performed according to the intention-to-treat (ITT) principle, defined as all included patients. PFS and OS definitions are available in Supplementary Appendix B (online only). Median followup was evaluated using the reverse KaplaneMeier method, and PFS and OS were estimated using the KaplaneMeier method and described with median and two-sided 95% CI. The database was looked for analysis on March 14, 2017. Treatments administered after the end of the protocol (8 cycles of FOLFIRINOX) were collected in the electronic case report form (eCRF): type of chemotherapy regimens, RT or CRT, rectal tumour resection and resection or ablation of metastases. All statistical analyses were conducted using SAS software 9.4 (SAS Institute, Cary, NC).

2.3. Central review Central review (O.L. and J.B.B.) of CT scans and rectal magnetic resonance imaging (MRI) was planned in the protocol and performed blinded to the clinical data and to the tumour response assessed by investigators. Tumour response according to RECIST criteria was assessed using CT scans, and response of rectal tumours was assessed by MRI. At baseline, after 4 and 8 cycles of FOLFIRINOX, tumour volumes were independently calculated according to the published methodology and using dedicated software (Myrian; Intrasense, Montpellier, France) [17]. A partial response was defined as a tumour volume reduction of at least 70%. 2.4. Study end-points The primary end-point was the 4-month disease control (4m DC) rate defined by the central review. The 4month DC rate was defined as the ratio of patients alive, evaluable and free of progression at primary and metastatic tumour sites at 4 months. All patients who received at least one dose of treatment and who had at

J.B. Bachet et al. / European Journal of Cancer 104 (2018) 108e116

111

3. Results Table 1 Patient clinical and laboratory characteristics. Characteristics

Total population n Z 65

Age, years Median Range Sex Male Female ECOG PS 0 1 2 Rectal tumoura Lower Middle Upper MD T stage at diagnosis (MRI)a T2 T3 T4 MD Differentiation grade Well Moderate Poor Unknown Mutational status RAS mutated BRAF mutated RAS/BRAF non-mutated Unknown Number of metastatic sites 1 2 Metastatic sites Liver metastases Liver metastases onlyb Lung metastases Nodes Peritoneal carcinomatosis Bone Adrenal Albumin (g/L) Median Range Bilirubinemia (x ULN) Median Range Creatinine clearance (mL/min) Median Range Alkaline phosphatase (x ULN) Median Range CEA (x ULN)  ULN > ULN and 5x ULN < 5x ULN

n Z 65 61 34e78 n Z 65 51 (78%) 14 (22%) n Z 65 24 (37%) 40 (61%) 1 (2%) n Z 65 22 (34%) 35 (54%) 5 (8%) 3 (4%) n Z 65 3 (4%) 13 (20%) 42 (65%) 7 (11%) n Z 65 23 (35%) 23 (35%) 3 (5%) 16 (25%) n Z 65 29 (53%) 3 (7%) 23 (35%) 10 (15%) n Z 65 24 (37%) 41 (63%) n Z 65 60 (92%) 21 (32%) 34 (52%) 18 (28%) 2 (3%) 2 (3%) 1 (2%) n Z 65 38.2 26.6e47.0 n Z 65 0.50 0.14e2.94 n Z 65 99.82 59.26e205.56 n Z 62 0.84 0.38e5.72 n Z 63 8 (13%) 10 (16%) 45 (71%)

CEA, carcinoEmbryonic antigen; ECOG PS, Eastern Cooperative Oncology Group performance status; MD, missing data; MRI, magnetic resonance imaging; ULN, upper limit of the normal. a MRI central review. b One patient had resectable metastases at inclusion (central review).

3.1. Patient characteristics From July 2012 to February 2015, 65 patients with rectal cancer and synchronous metastases were included in the study and treated in 22 participating French centres (Supplementary Fig. A.1). Patient characteristics are described in Table 1. 3.2. Compliance with induction treatment All patients received at least one cycle of FOLFIRINOX, and 56 (85%) received the 8 planned cycles of induction (median 8; range 1e8). The median treatment duration was 3.48 months (range 0.03e6.70). Forty-six patients (71%) had at least one dose reduction, and 48 patients (74%) had at least one delayed dose. The median dose intensities (all cycles) were 93.5% for oxaliplatin, 96.5% for irinotecan, 86% for 5-fluorouracil (5FU) bolus and 96% for continuous 5FU infusion. Forty-six patients (71%) received G-CSF, in most cases (87%) as primary prophylaxis. 3.3. Primary objective and ORRs At an interim efficacy analysis in April 2014, the 4-month DC rate was 92% (95% CI, 76.9e98.6) (n Z 23/25), and Table 2 Objective response rates. Response rate (RECIST 1.1)

After 4 cycles

After 8 cycles

Investigator review Partial response Stable disease Progressive disease Non-evaluable Central independent review Partial response Stable disease Progressive disease Non-evaluable Response rate of rectal tumour (MRI, volume) CD not recoveredc Central independent review Partial response Stable disease Progressive disease Non-evaluable

n Z 64 29 (45%) 31 (48%) 3 (5%) 1 (2%) n Z 64 30 (47%)a 29 (45%) 2 (3%) 3 (5%) After 4 cycles n Z 64 17 (27%) n Z 47 23 (49%) 24 (51%)b 0 0

n Z 64 49 (77%) 9 (14%) 6 (9%) 0 n Z 64 55 (86%) 5 (8%) 4 (6%)a 0 After 8 cycles n Z 64 16 (25%) n Z 48 30 (63%) 17 (35%) 0 1 (2%)b

CDCompact Disc; MRI, magnetic resonance imaging; RECIST, Response Evaluation Criteria in Solid Tumours. a One patient with a partial response at central independent review was considered progressive by the investigator (appearance of a new lesion, not confirmed at centralised review) and thus was considered in progression after 8 cycles. b One patient with stable disease at MRI after 4 cycles discontinued the study because the investigator reported clinical local progression. c MRI CD rom were not available for 17 patients and 16 patients for the radiologic evaluations after 4 cycles and 8 cycles, respectively.

112

J.B. Bachet et al. / European Journal of Cancer 104 (2018) 108e116

recruitment was pursued to include the expected 65 patients. Final analysis was conducted as planned on the mITT population (64 evaluable patients). One patient was excluded because of death unrelated to study treatment (stroke) after the first cycle (Supplementary Fig. A.1). The 4-month DC rate was 91% (95% CI, 82.3e95.8) according to the investigators and 94% (95% CI, 86.3e97.8) according to the review committee. The ORRs are summarised in Table 2 and Fig. 1.

eight patients (58%) had RT or CRT for the rectal tumour, 32 patients (49%) had resection of the rectal tumour and 22 patients (34%) had at least one surgery and/or ablation of metastases. Among the 21 patients with liver metastases only, 13 patients (59%) had metastases resection and/or thermal ablation combined with rectal resection for 9 of them (after RT or CRT in 5 patients) and with RT or CRT without rectal resection for 4 of them. Treatments received after the end of the study treatment are described in Table 3.

3.4. Efficacy on local symptoms 3.6. Survival At inclusion, 47 patients (72%) had local symptoms: rectal bleeding (n Z 27, 42%) and/or rectal syndrome (n Z 39, 60%). Local symptoms, rectal bleeding and/or rectal syndrome were seen in 16%, 5% and 14% of patients after 4 cycles of FOLFIRINOX, and in 10%, 3% and 7%, after 8 cycles, respectively. No baseline or ontreatment obstructive symptoms were observed. 3.5. Treatment after induction treatment After the 8 planned cycles of FOLFIRINOX, 60 patients (92%) continued systemic chemotherapy. Thirty-

After a median follow-up of 35.0 months (95% CI, 31.3e41.7), progression occurred in 60 patients (92%), and 41 patients (63%) were dead. In ITT (n Z 65), median PFS was 10.9 months (95% CI, 8.8e12.9), and median OS was 33.4 months (95% CI, 22.6e38.2) (Fig. 2). Median PFS and OS were 13.1 (95% CI, 8.8e18.9) and 42.1 (95% CI, 33.4-not reached) months for the 22 patients (34%) who had secondary resection of metastases, 11.3 (95% CI, 8.1e13.4) and 38.8 (95% CI, 13.6e44.6) months for the 15 patients (23%) who had secondary resection of the primary tumour but not of

Fig. 1. Water fall plots of objective response rates assessed by independent central review. Objective response rates according to RECIST criteria were assessed on chest-abdominal-pelvic CT scans blinded to clinical data (A) after 4 cycles of FOLFIRINOX (62 evaluable patients) and (B) after 8 cycles of FOLFIRINOX (63 evaluable patients). Objective response rates of the rectal tumour were assessed by MRI (volume, partial response defined by a decrease  70%) (C) after 4 cycles of FOLFIRINOX (47 evaluable patients) and (D) after 8 cycles of FOLFIRINOX (47 evaluable patients). CT, computed tomography; MRI, magnetic resonance imaging.

J.B. Bachet et al. / European Journal of Cancer 104 (2018) 108e116

113

Table 3 Description of treatments received after the end of 8 planned cycles of FOLFIRINOX. Type of treatment

n Z 65

Chemotherapy Drugs used during follow-up Fluoropyrimidine Irinotecan Oxaliplatin Bevacizumab Aflibercept Cetuximab Panitumumab Others Number of chemotherapy linesa One Two Three Four Oxaliplatin re-introduction Local treatment of rectal tumour Radiotherapy (RT) Chemoradiotherapy (CRT) RT or CRT without rectal resection Rectal resection After RT or CRT Without RT or CRT Proctectomy Abdominoperineal amputation Type of resection R0 R1 Local treatment of metastases Metastasis resection (MR) One MR Two MR Liver metastases thermal ablation

60 (92%) 60 (92%) 54 (83%) 36 (55%) 34 (52%) 7 (11%) 14 (22%) 8 (12%) 12 (18%) 12 (18%) 27 (42%) 19 (29%) 2 (3%) 23 (35%) 38 (58%)b 27 (42%) 12 (18%) 15 (23%) 32 (49%) 23 (35%) 9 (14%) 26 (40%) 6 (9%) 29 (44%) 3 (5%) 22 (34%) 20 (31%) 15 (23%) 5 (8%) 9 (14%)

a

A line was defined by the first administration of a new drug; thus, administration of 5-fluorouracil, oxaliplatin and irinotecan during other lines was not considered as a new line in the absence of others associated drugs. For this reason, the re-introduction of oxaliplatin has been individualised. b One patient received radiotherapy and then chemoradiotherapy for local rectal progression.

metastases and 7.4 (95% CI, 52e11.7) and 17.6 months (95% CI, 11.8e26.4) for the 28 patients (43%) who had no secondary resection (primary tumour or metastases), respectively. Details of oncologic outcomes according to RAS and BRAF status are summarised in Table 4. 3.7. Safety The rates of side-effects according to NCI CTCAE, version 4.0, are reported in Table 5. Twenty-three serious adverse events were reported in 14 patients (22%). No toxic death related to study treatment was reported. 4. Discussion This multicenter phase II study showed that induction chemotherapy with FOLFIRINOX is feasible and active

Fig. 2. Progression-free survival and overall survival using KaplaneMeier curves for the 65 patients in the intention-to-treat population.

on primary and metastatic tumour sites in patients with rectal cancer and synchronous unresectable metastases. The trial met the pre-specified primary end-point with a 4-month DC rate above 90%. The toxicity profile was safe and acceptable. The best strategy for treating rectal cancer with synchronous unresectable metastases is not consensual [1]. In such patients, the main goal is not only to control the metastatic disease, which may rapidly impair survival, but also to control the rectal tumour and prevent severe local symptoms from occurring and major degradation of patients’ quality of life. Few prospective trials have assessed therapeutic options in this setting, and, to our knowledge, the FFCD 1102 trial is the first phase II trial to assess the efficacy and tolerability of upfront chemotherapy. When metastatic, rectal tumours are most often locally advanced and symptomatic. Upfront rectal surgery has the advantage of immediately controlling local symptoms, especially pain [6,18]. Nevertheless, resection may be incomplete in the absence of pre-operative treatment and carries a high risk of local recurrence. Upfront surgery may also be associated with postoperative mortality, ranging from 0% to 11%, and major post-operative complications, ranging from 4% to 33%, sometimes with a rapid change in the patients’ general condition that will delay or prevent the initiation of chemotherapy and negatively impact survival [6,8e10,19]. Upfront rectal surgery is thus no longer recommended [1]. Three options are generally discussed: (1) RT with concomitant systemic chemotherapy, (2) short-course RT (5  5) followed by rapid initiation of systemic chemotherapy and (3) upfront systemic chemotherapy.

114

J.B. Bachet et al. / European Journal of Cancer 104 (2018) 108e116

Table 4 Oncologic outcome according to RAS and BRAF status.

Partial response after 8 cycles a Disease control after 8 cycles a Progression-free survival, median (95% CI) Overall survival, median (95% CI)

RAS and BRAF non-mutated n Z 23

RAS mutated n Z 29

BRAF mutated n Z 3

Unknown n Z 10

100% 100% 11.30 (9.66e16.69) 37.82 (26.42eNR)

79% 93% 11.33 (6.21e14.88) 32.89 (12.48e38.24)

50%b 50%b 1.74 (1.05e13.54) 4.27 (1.05e26.55)

80% 90% 8.56 (3.65e11.70) 19.35 (11.56e44.62)

CI, confidence interval; NR, not reached; OS, overall survival; PFS, progression-free survival. a Central independent review. b Only two patients were evaluable.

CRT completely resolves symptoms in 94% of patients who have a symptomatic rectal tumour at diagnosis, but the period in which resection after CRT can be performed is limited, and the dose intensity of the chemotherapy used is non-optimal for the control of metastases [7]. In patients with symptomatic unresectable metastases, French recommendations were to combine CRT and systemic chemotherapy, but such a therapeutic strategy is not easy to manage and to put in place, and few centres use it [10,20]. Upfront short-course RT followed by immediate systemic chemotherapy is an attractive option because it could allow good control of pelvic symptoms without delaying chemotherapy. The efficacy of short-course RT in metastatic rectal cancer has been evaluated in 18 patients with obstructive symptoms pre-treated or not with chemotherapy [21]. Its efficacy on local symptoms was substantial, and the authors concluded that shortcourse RT is a valuable option in avoiding colostomy. Table 5 Safety of FOLFIRINOX induction therapy (NCI CTCAE criteria v4.0). Toxicity

Grade 1

At least one Hematologic Anaemia Thrombocytopenia Neutropenia (without fever) Febrile neutropenia At least one haematologic Non-hematologic Nausea Vomiting Diarrhoea Mucositis Fatigue Hand foot syndrome Paraesthesia Alopecia Fever without neutropenia Thromboembolic event Increased ALAT Increased ASAT Increased bilirubin

65 (100%) 63 (97%) 44 (68%) 9 (14%)

Grade 2 Grade 3 Grade 4

54 (83%) 1 (2%) 28 (43%) e 55 (85%)

17 (26%) 1 (2%) 20 (31%) e 19 (29%)

0 0 14 (22%) 2 (3%) 2 (3%)

0 0 5 (8%) 0 0

31 (48%) 17 (26%) 48 (74%) 18 (28%) 52 (80%) 5 (8%) 46 (71%) 18 (28%) 10 (15%) 4 (6%) 20 (31%) 23 (35%) 3 (5%)

16 (25%) 6 (9%) 14 (22%) 3 (5%) 28 (43%) 0 19 (29%) 13 (20%) 2 (3%) 3 (5%) 4 (6%) 3 (5%) 1 (2%)

3 (5%) 0 8 (12%) 1 (2%) 5 (8%) 0 1 (2%) e 0 2 (3%) 1 (2%) 1 (2%) 0

0 0 0 1 (2%) 0 0 0 e 0 0 0 0 0

ALAT, alanine transaminase; ASAT, aspartate aminotransferase; NCI CTCAE, National Cancer Institute Common Terminology Criteria for Adverse Events.

However, the value of upfront short-course RT remains to be assessed because it may make secondary rectal resection more difficult and therefore impair possible future therapeutic options in the event of a good response to chemotherapy. Upfront chemotherapy is also debated, in this case because of its potential inability to control primary tumour symptoms as compared to surgery or CRT. An advantage of starting treatment with systemic chemotherapy is that it does not close any therapeutic options, which may all be secondarily considered depending on the response to treatment of primary tumour and metastases and the patient’s condition. Moreover, recent data argue for satisfying symptomatic control and primary tumour regression in patients with rectal cancer treated with systemic chemotherapy [13e15]. In these studies conducted in patients with non-metastatic locally advanced rectal cancer, a combination of oxaliplatin and fluoropyrimidine, with or without bevacizumab, was associated with local tumour control close to 100% and a decrease or disappearance of local symptoms in at least 90% of patients [12e14]. Our results obtained with the FOLFIRINOX regimen are of the same order, and we report for the first time a precise assessment of the rectal tumour response by volumetric MRI measurement. Only one patient experienced local symptoms increase during the induction period. This patient discontinued the study because of rectal pain, diarrhoea and suspicion of local progression. However, the rectal tumour was assessed as stable on MRI after 4 cycles (central review). Considering the whole study population, after 8 planned cycles of chemotherapy, only 10% of patients still had rectal symptoms vs 72% at baseline. When this study was designed, the triplet chemotherapy regimen (FOLFOXIRI or FOLFIRINOX) was the regimen associated with the highest ORR [15,22] Moreover, the risk of initial progression described with this regimen was low (6%e11%) and two times inferior to those reported with the FOLFOX or FOLFIRI (12%e24%) [15,22e24]. Since then, high ORRs and low initial progression rates have also been described with others regimens combining doublet chemotherapy and anti-EGFR antibody for RAS wild-type tumours subgroup or FOLFOXIRI plus bevacizumab [24e26]. Based on our results, we can expect that comparable

J.B. Bachet et al. / European Journal of Cancer 104 (2018) 108e116

results could be obtained with these regimens. However, because of the lack of data of triplet chemotherapy plus a targeted agent when the present study was designed and the theoretical risk of local complications (tumour perforation and fistulae) with bevacizumab, we opted for the FOLFIRINOX-alone regimen. The therapeutic strategy evaluated in our trial, 8 cycles of FOLFIRINOX followed by a free therapeutic strategy defined by the investigators, also leads to interesting PFS and OS results. These median survivals are very close to those reported in the TRIBE phase III trial in metastatic colon cancer patients with FOLFOXIRI þ bevacizumab [16]. In our study as in others, patients who had secondary resections had the best median OS (42 months), confirming that surgical removal of all primary and metastatic lesions improves outcomes in metastatic rectal cancer patients as in colon cancer [11].

115

Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.ejca.2018.09.006. Appendix A1. List of investigators who participated in the study and not cited among the authors of the article: L. Dahan (CHU La Timone, Marseille, France) D. Auby (CH Layne´, Mont de Marsan, France) F. Khemissa (CH St Jean, Perpignan, France) F. Ghiringhelli (CAC GF. Leclerc, Dijon, France) S. Nguyen (CH Beauvais, France) A. Bedjaoui (CH du Leman, Thonon les Bains, France) E. Terrebonne (CHU Haut-Le´ve`que, Pessac, France) J. Thaury (CH, Pau, France) M. Baconnier (CH re´gional, Annecy-Pringy, France)

5. Conclusion

References In conclusion, in patients with rectal cancer with synchronous unresectable metastases, induction chemotherapy with FOLFIRINOX is effective on both metastases, which dominate the prognosis, and on the primary rectal tumour and its relative symptoms. After induction chemotherapy, the therapeutic strategy can be adapted according to treatment side-effects, the patient’s condition, symptoms, tumour response and the possibility of surgical resection of the primary rectal tumour and metastatic lesions, leading to a personalised optimal treatment strategy for each patient. Funding The FFCD Group sponsored the trial, and investigators were responsible for study design, data collection, data analysis and data interpretation. The writing of the report and the decision to submit for publication were the responsibility of the FFCD Cooperative Group. Conflict of interest statement J.B.B. has received personal fees from Amgen, Bayer, Celgene, Merck Serono, Roche, Sanofi and Servier and non-financial support from Amgen, Merck Serono and Roche. J.L.R. has received personal fees from Merck Serono, Bayer and BTG. J.T. has received personal fees from Lilly, Amgen, Celgene, Merck Serono, Servier, Shire, Sanofi, Sirtex and Roche. All the other authors declare no competing interests.

Acknowledgements The authors thank all the investigators and research technicians who helped them to carry out this study. The authors thank La Ligue contre le cancer for its support.

[1] Lutz MP, Zalcberg JR, Glynne-Jones R, et al. Second St. Gallen European Organisation for research and treatment of cancer gastrointestinal cancer conference: consensus recommendations on controversial issues in the primary treatment of rectal cancer. Eur J Cancer 2016;63:11e24. [2] Bosset JF, Collette L, Calais G, et al. Chemotherapy with preoperative radiotherapy in rectal cancer. N Engl J Med 2006; 355(11):1114e23. [3] Kapiteijn E, Marijnen CAM, Nagtegaal I, et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001;345(9):638e46. [4] Manceau G, Brouquet A, Bachet JB, et al. Response of liver metastases to preoperative radiochemotherapy in patients with locally advanced rectal cancer and resectable synchronous liver metastases. Surgery 2013;154(3):528e35. [5] Yoon HI, Koom WS, Kim TH, et al. Upfront systemic chemotherapy and short-course radiotherapy with delayed surgery for locally advanced rectal cancer with distant metastases: outcomes, compliance, and favorable prognostic factors. PLoS One 2016; 11(8), e0161475. [6] Nash GM, Saltz LB, Kemeny NE, et al. Radical resection of rectal cancer primary tumor provides effective local therapy in patients with stage IV disease. Ann Surg Oncol 2002;9(10): 954e60. [7] Crane CH, Janjan NA, Abbruzzese JL, et al. Effective pelvic symptom control using initial chemoradiation without colostomy in metastatic rectal cancer. Int J Rad Oncol Biol Phys 2001;49(1): 107e16. [8] Benoist S, Pautrat K, Mitry E, Rougier P, Penna C, Nordlinger B. Treatment strategy for patients with colorectal cancer and synchronous irresectable liver metastases. Br J Surg 2005;92(9): 1155e60. [9] Scheer MGW, Sloots CEJ, van der Wilt GJ, Ruers TJ. Management of patients with asymptomatic colorectal cancer and synchronous irresectable metastases. Ann Oncol 2008;19(11): 1829e35. [10] Benoist S. Recommendations for clinical practice. Therapeutic choices for rectal cancer. How should rectal cancers with synchronous metastases be managed? Gastroenterol Clin Biol 2007; 31(Spec No 1). 1S75-80, S100-2. [11] Van Cutsem E, Cervantes A, Adam R, et al. ESMO consensus guidelines for the management of patients with metastatic colorectal cancer. Ann Oncol 2016;27(8):1386e422.

116

J.B. Bachet et al. / European Journal of Cancer 104 (2018) 108e116

[12] Chua YJ, Barbachano Y, Cunningham D, et al. Neoadjuvant capecitabine and oxaliplatin before chemoradiotherapy and total mesorectal excision in MRI-defined poor-risk rectal cancer: a phase 2 trial. Lancet Oncol 2010;11(3):241e8. [13] Fernandez-Martos C, Brown G, Estevan R, et al. Preoperative chemotherapy in patients with intermediate-risk rectal adenocarcinoma selected by high-resolution magnetic resonance imaging: the GEMCAD 0801 Phase II Multicenter Trial. Oncol 2014; 19(10):1042e3. [14] Schrag D, Weiser MR, Goodman KA, et al. Neoadjuvant chemotherapy without routine use of radiation therapy for patients with locally advanced rectal cancer: a pilot trial. J Clin Oncol 2014;32(6):513e8. [15] Falcone A, Ricci S, Brunetti I, et al. Phase III trial of infusional fluorouracil, leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) compared with infusional fluorouracil, leucovorin, and irinotecan (FOLFIRI) as first-line treatment for metastatic colorectal cancer: the Gruppo Oncologico Nord Ovest. J Clin Oncol 2007;25(13):1670e6. [16] Cremolini C, Loupakis F, Antoniotti C, et al. FOLFOXIRI plus bevacizumab versus FOLFIRI plus bevacizumab as first-line treatment of patients with metastatic colorectal cancer: updated overall survival and molecular subgroup analyses of the openlabel, phase 3 TRIBE study. Lancet Oncol 2015;16(13):1306e15. [17] Nougaret S, Rouanet P, Molinari N, et al. MR volumetric measurement of low rectal cancer helps predict tumor response and outcome after combined chemotherapy and radiation therapy. Radiology 2012;263(2):409e18. [18] Tougeron D, Di Fiore F, Lefebure B, et al. Control of pelvic symptoms in patients with rectal cancer and synchronous metastases. Gastroenterol Clin Biol 2009;33(12):1106e13.

[19] Smith JD, Butte JM, Weiser MR, et al. Anastomotic leak following low anterior resection in stage IV rectal cancer is associated with poor survival. Ann Surg Oncol 2013;20(8): 2641e6. [20] Michael M, Chander S, McKendrick J, et al. Phase II trial evaluating the feasibility of interdigitating folfox with chemoradiotherapy in locally advanced and metastatic rectal cancer. Br J Cancer 2014;111(10):1924e31. [21] Picardi V, Deodato F, Guido A, et al. Palliative short-course radiation therapy in rectal cancer: a phase 2 study. Int J Radiat Oncol Biol Phys 2016;95(4):1184e90. [22] Masi G, Allegrini G, Cupini S, et al. First-line treatment of metastatic colorectal cancer with irinotecan, oxaliplatin and 5fluorouracil/leucovorin (FOLFOXIRI): results of a phase II study with a simplified biweekly schedule. Ann Oncol 2004;15(12): 1766e72. [23] Tournigand C, Andre´ T, Achille E, et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol 2004;22(2):229e37. [24] Douillard JY, Siena S, Peeters M, Koukakis R, Terwey JH, Tabernero J. Impact of early tumour shrinkage and resection on outcomes in patients with wild-type RAS metastatic colorectal cancer. Eur J Cancer 2015;51(10):1231e42. [25] Loupakis F, Cremolini C, Masi G, et al. Initial therapy with FOLFOXIRI and bevacizumab for metastatic colorectal cancer. N Engl J Med 2014;371(17):1609e18. [26] Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic colorectal cancer (FIRE3): a randomised, open-label, phase 3 trial. Lancet Oncol 2014; 15(10):1065e75.