Phase II study on first-line treatment of NIVolumab in combination with folfoxiri/bevacizumab in patients with Advanced COloRectal cancer RAS or BRAF mutated – NIVACOR trial (GOIRC-03-2018)

Background FOLFOXIRI (fluorouracil, leucovorin, oxaliplatin, and irinotecan) plus bevacizumab has shown to be one of the therapeutic regimens in first line with the highest activity in patients (pts.) with metastatic colorectal cancer (mCRC) unselected for biomolecular alterations. Generally, tumors co-opt the programmed death-1/ligand 1 (PD-1/PD-L1) signaling pathway as one key mechanism to evade immune surveillance. As today, anti-PD-1 monoclonal antibodies are FDA approved only for DNA mismatch repair deficient/microsatellite instability-high (MMRd/MSI-H), which represent only about 5% among all mCRC. Nowadays, there are no data demonstrating anti PD-1 activity in proficient and stable disease (MMRp/MSS). A different target in mCRC is also the Vascular Endothelial Growth Factor A (VEGF-A), which acts on endothelial cells to stimulate angiogenesis. VEGF-A inhibition with bevacizumab has shown to increase the immune cell infiltration, providing a solid rationale for combining VEGF targeted agents with immune checkpoint inhibitors. Based on these evidences, we explore the combination of triplet chemotherapy (FOLFOXIRI) with bevacizumab and nivolumab in pts. with mCRC RAS/BRAF mutant regardless of microsatellite status. Methods/design This is a prospective, open-label, multicentric phase II trial where pts. with mCRC RAS/BRAF mutated, in first line will receive nivolumab in combination with FOLFOXIRI/bevacizumab every 2 weeks for 8 cycles followed by maintenance with bevacizumab plus nivolumab every 2 weeks. Bevacizumab will be administered intravenously at dose of 5 mg/kg every 2 weeks and nivolumab intravenously as a flat dose of 240 mg every 2 weeks. The primary endpoint is the overall response rate (ORR). This study hypothesis is that the treatment is able to improve the ORR from 66 to 80%. Secondary endpoints include OS, safety, time to progression, duration of response. Collateral translational studies evaluate the i) tumor mutational burden, and ii) genetic alterations by circulating free DNA (cfDNA) obtained from plasma samples. The trial is open to enrollment, 9 of planned 70 pts. have been enrolled. Trial registration NIVACOR is registered at ClinicalTrials.gov: NCT04072198, August 28, 2019.


Background
The colorectal cancer (CRC) is diagnosed at advanced stages in almost 50%, and in this setting the 5-year survival rate is approximately of the 12% [1]. CRC is a heterogeneous tumor consisting of multiple genetic, genomic and epigenetic alterations, and this entails the stratification of the patients into different subgroups susceptible to different treatments. In 2015, a large-scale consortium reported four consensus molecular subtypes (CMS) of CRC described in MSI Immune (CMS1), Canonical (CMS2), Metabolic (CMS3), and Mesenchymal (CMS4) [2,3], each with specific biomolecular and prognostic features.
In metastatic CRC (mCRC) the RAS gene (KRAS, NRAS) is mutated approximately in 50-55%. Currently, detection of RAS mutations is the only predictive marker of response to the anti-EGFR antibodies, cetuximab and panitumumab [4,5].
The second potential biomarker in mCRC is BRAF, mutated in 5-11% of cases [6]. The BRAFV600E point mutation is the most common alteration and believed to be mutually exclusive with KRAS exon 2 mutations [7]. Accordingly, several clinical trials have highlighted the negative prognostic role of the BRAF mutation associated with high mortality [8].
In patients who harbor RAS/BRAF mutant tumors, the addition of anti-vascular growth factor (VEGF) antibody to cytotoxic drugs based on fluorouracil/levofolinate/irinotecan or oxaliplatin, has become one of the standard treatments in first-line of mCRC [9].
Several randomized studies, have proved that the triplet of chemotherapy with fluorouracil/levofolinate/irinotecan/ oxaliplatin (FOLFOXIRI) combined to bevacizumab is more effective than doublet of chemotherapy plus bevacizumab, and this combination was well tolerated as firstline treatment in selected fit patients [10,11]. In the TRIBE study [9], a phase III study, in first-line setting the treatment with FOLFOXIRI plus bevacizumab improved the primary endpoint, progression-free survival (PFS), compared with FOLFIRI (fluorouracil, leucovorin, and irinotecan) plus bevacizumab (HR 0.75; 95% CI 0.62-0.90; p = 0.003). A significant improvement and depth of tumor response associated with early tumor shrinkage assessed by Response Evaluation Criteria In Solid Tumors (RECI ST) version 1.0, was also reported in experimental arm (FOLFOXIRI plus bevacizumab). Moreover, an advantage in terms of median overall survival (mOS) in FOLFOXIRI plus bevacizumab arm was revealed (29.8 months vs. 25.8 months; HR 0.80, 95% CI 0.65-0.98; p = 0.03). The molecular sub-analysis of the TRIBE study showed a better mOS in RAS and BRAF wild-type subgroups compared to RAS mutated and BRAF mutated subgroups (37.1 months vs. 25.6 months vs. 13.4 months), respectively [12].
Currently, a further tumor feature being studied and of the great interest is the description of immune landscape of the microenvironment in mCRC, especially concern to microsatellite status. Most of tumors (85-90%) had a low-to-moderate mutation load and two main groups of CRCs were recognized: proficient in terms of mismatch repair mechanisms (MMRp) of DNA and microsatellite-stable (MSS). The minority are highly mutated with deficient mismatch-repair mechanisms (MMRd) relating to a microsatellite-instable phenotype (MSI, more accurately MSI-high [MSI-H]). This classification systems, MMRp/MSS vs. MMRd/MSI afford a way to stratify patients concerning to immunotherapy response [15,16]. Nowadays, the immunotherapy approach with anti-PD1/PD-L1 antibodies for mCRC has demonstrated efficacy only in MMRd/MSI tumor subgroups but no in MMRp/MSS tumors. Several phase I-II-III clinical trials have been conducted (Table 1) and others are still ongoing (Table 2) to establish the immunotherapeutic efficacy alone or in combination with  other drugs, especially with chemotherapy. Preclinical studies, have shown the close connection between tumor cells and the tumor microenvironment (TME) status, especially the surrounding milieu composed by the stroma, tumor-infiltrating lymphocytes, and lymphatic and vascular layers. In this context, endothelial cells play a key role in the extravasation of immune cells, influencing the arrangement of the tumor environment [38]. It is known that extremely inflamed tumors reflect poor tumor angiogenesis; however, highly vascularized tumors may conversely entail tumors with deprived immune infiltration. One of the biomolecules responsible for affecting the hematopoietic progenitor cell differentiation to dendritic cells (DCs) is the tumor-derived VEGF. DCs are the most efficient antigen presenting cells due to the peptide presentation of tumor antigens on the major histocompatibility complex (MHC) I and II molecules, eliciting Tcells by B7 molecule expression, against cancer antigens [39,40]. Active extravasation of leukocytes in the tumor stroma requires a series of events starting from the rolling, firm adhesion of leukocytes on endothelial cells, and leading to wandering into the interstitial areas. VEGF plays an essential role in this process as the blood vessels could present an obstacle to extravasation of immune cells in the interstitial space [41][42][43].
Additionally, VEGF inhibition by bevacizumab, involves a normalization of tumor vascularization rises the permeability to immune cell infiltration.
It is essential to restore an immunological environment to sensitize mCRC to immune checkpoint inhibitors, to combine them with treatments that stimulate T-cells as chemotherapy, although the molecular mechanisms of sensitization are still not clear. Preclinical models suggest that some chemotherapies can improve the immunotherapy efficacy [52,53]. The association of fluorouracil and oxaliplatin with immune checkpoint inhibitors in vivo could deplete Myeloid-derived Suppressor Cells (MDSCs) [54], and trigger an immunogenic arrangement of tumor cell death [55]. Dosset et al. [56], have investigated in two mouse models the use of FOLFOX in association with anti-PD-1 therapy. The combination induced a strong expression of PD-1 on CD8 + TILs, and the IFN-γ secreted by FOLFOX-induced CD8 + T cells leads PD-L1 expression on tumor cells and this mechanism is considered as an adaptive immune resistance system to FOL-FOX. In neoadjuvant setting, mCRC patients treated with FOLFOX showed an increased CD8 + cell infiltrate and tumor PD-L1 expression. Another chemotherapeutic drug, trifluridine/tipiracil (FTD/TPI), an antimetabolite agent used to treat chemo-refractory mCRC, induced immunogenic arrangement of tumor cell death in vitro in MSS CT26 mouse colon carcinoma cell line, as well as in various human MSS colorectal cancer cell lines [57]. In vivo, the combination of FTD/TPI with oxaliplatin was able to induce immunogenic arrangement of tumor cell death, but not the single agents. Furthermore, the combination abolished type-2 tumor-associated macrophages (TAM2), resulting in higher cytotoxic CD8 + T-cell infiltration and activation. This effect was associated with tumor cells PD-L1 expression and PD-1 induction in CD8 + T cells, resulting in T-cell exhaustion. Based on these preclinical and clinical data, there is sufficient evidence to explore the combination of chemotherapy with immunotherapy and antiangiogenetic inhibitors in pts. with mCRC RAS/ BRAF mutated.

Protocol overview/study treatment
This is a prospective, open-label, multicentric phase II trial in which pts. with RAS or BRAF mutated will receive nivolumab in combination with FOLFOXIRI/bevacizumab as first line treatment. Study screening will take place within 28 days prior to initiation of study treatment. At screening, every patient must have local RAS/ BRAF known status. A centralized review of RAS/BRAF status will be performed. Eligible pts. will be enrolled and begin treatment with FOLFOXIRI/bevacizumab plus nivolumab every 2 weeks for 8 cycles followed by maintenance with bevacizumab plus nivolumab every 2 weeks until disease progression, unacceptable toxicity or patient/physician decision. Bevacizumab will be administered intravenously at dose of 5 mg/kg every 2 weeks. Nivolumab will be administered intravenously at flat dose of 240 mg every 2 weeks. FOLFOXIRI will be administered as 165 mg/m 2 intravenous infusion of irinotecan for 60 min, followed by an 85 mg/m 2 intravenous infusion of oxaliplatin given concurrently with leucovorin at a dose of 200 mg/m 2 for 120 min, followed by a 3200 mg/m 2 continuous infusion of fluorouracil for 48 h (Fig. 1).
During the protocol's treatment, pts. will be followed for safety based on Adverse Event (AE) assessments including vital signs, physical findings and clinical laboratory test results.
In order to guarantee the safety of pts., the enrolment will be stopped when the 10th patient will start treatment. An Independent Monitoring Committee will evaluate the safety data of these pts. and will decide if the study should be completed, amended or closed.
The efficacy will be evaluated by the investigator according to RECIST 1.1 criteria every 8 weeks during treatment, and then every 3 months for 3 years.
During the study baseline tumor blocks will be centrally analyzed to determinate MSI/MSS and PD-L1 status, inflammatory infiltrate through evaluation of high peri-and/or intra-tumor lymphocyte infiltration (TIL) using CD3 + and CD8 + TILs, tumor-associated macrophages (TAMs), tumor-associated neutrophils (TANs), and regulatory T cells (Tregs) well as the expression of marker of autophagy.
Following discontinuation of the treatment, safety assessments will be conducted 30 days after the last drug administration or until initiation of other anti-cancer therapy. Thereafter, pts. will be followed for disease progression (unless this has already occurred), serious AEs, anticancer therapy and survival. Follow-up will continue for up to 3 years.
A blood sample will be collected at baseline, prior to cycle 5, at the end of chemotherapy and at disease progression. Quality of life will be assessed at baseline, every 4 weeks during treatment and study discontinuation visit.
A list of participating centers is provided in Table 3.

Inclusion criteria
For inclusion in the study, all of the following inclusion criteria must be fulfilled: (i) histopathological confirmed colon adenocarcinoma; (ii) initially unresectable metastatic colorectal cancer not previously treated with chemotherapy for metastatic disease; (iii) assessment of RAS and BRAF status of the primary and/or secondary colon cancer on biopsies (mutant); (iv) age ≥ 18 years and ≤ 75 years; (v) ECOG performance status 0-1; (vi) if dihydropyridine dehydrogenase (DPD) status is known it must be wild type; (vii) laboratory data including: white blood cell count ≥3 × 10 9 /L with neutrophils ≥1.5 × 10 9 / L, platelet count ≥100 × 10 9 /L, hemoglobin ≥9 g/dL (5,6 mmol/l), total bilirubin ≤1.5 x ULN (upper limit of normal), ASAT and ALAT ≤2.5 x ULN, alkaline phosphatase ≤1.5 x ULN, serum creatinine ≤1.5 x ULN; (viii) signed written informed consent obtained prior to any study specific screening procedures.

Exclusion criteria
Patients are not eligible for this study if any of the following exclusion criteria apply: (i) prior chemotherapy, excluded pts. treated in neo/adjuvant setting at least 12 months before diagnosis of metastatic disease; (ii) radiotherapy to any site within 4 weeks before the study; (iii) evidence of bleeding diathesis or coagulopathy; (iv) uncontrolled hypertension and prior history of hypertensive crisis or hypertensive encephalopathy (v) systemic corticosteroids within 2 weeks of the first dose of nivolumab; (vi) diagnosis of immunodeficiency or is receiving systemic steroid therapy within 14 days prior to the first dose of trial treatment; (vii) active and untreated brain (CNS) metastases and/or carcinomatous meningitis or subjects with previously treated brain metastases may participate provided they are not using steroids for at least 7 days prior to trial treatment; (viii) evidence of interstitial lung disease, active non-infectious pneumonitis, or a history of grade 3 or greater pneumonitis; (ix) live vaccine within 30 days prior to the first dose of trial treatment; (x) significant vascular disease (e.g. aortic aneurysm requiring surgical repair or recent arterial thrombosis) within 6 months of study enrollment; (xi) history of abdominal fistula, gastrointestinal (GI) perforation, intra-abdominal abscess or active GI bleeding within 6 months prior to the first study treatment; (xii) pregnancy (absence to be confirmed by ß-hCG test) or breast-feeding period; (xiii) any significant disease which, in the investigator's opinion, would exclude the patient from the study.

Study endpoints
The present trial will determine if adding nivolumab to the first line therapy with FOLFOXIRI/bevacizumab is efficient in terms of response rate in mCRC RAS/BRAF mutated. To evaluate the Overall Response Rate (ORR), defined as complete response (CR), partial response (PR), and stable disease (SD), we will use RECIST version 1.1 criteria. Secondary endpoints are the following: (i) safety assessment of the combination treatment with FOLFOXIRI/ bevacizumab plus nivolumab graded by National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE) v. 4.03; (ii) OS defined as the time from beginning of the study-drug administration to the date of death from any cause; (iii) Time To Progression (TTP) defined as the time from beginning of the studydrug administration and the first date of documented progression, based on investigator assessment as per RECIST 1.1 criteria, or death due to any cause, whichever occurs first; (iv) the duration of response defined as the time between the first evidence of response (SD/PR/CR) and the date of documented progression or death due to any cause; (v) the quality of life of pts. determinate with the EORTC QLQ-C30 that consists of 30 questions that assess five aspects of patient functions (physical, emotional, role, cognitive, and social), three symptom scales (fatigue, nausea and vomiting, pain), global health and/or quality of life, and six single items (dyspnea, insomnia, appetite loss, constipation, diarrhea, financial difficulties) with a recall period of the previous week. Scale scores can be obtained for the multi-item scales.
The collateral study includes the TMB, MSI status and the role of genetic and molecular pattern analysis in relation to patient's outcome. Formalin-fixed and paraffinembedded (FFPE) tumor samples will be collected before starting fist-line therapy (at baseline), as primary and/or metastatic tumor tissue blocks or as 15 5-μm unstained slides. The neoplastic cell content of each tumor sample will be assessed and in those cases with neoplastic cells < 50% a macro-dissection of the specimen will be performed, if possible. For all the pts. enrolled, venous blood will be obtained by standard phlebotomy technique from a peripheral access point or from a central line, by trained personnel. Blood samples will be collected at different points: at baseline, prior to 5 cycle, at the end of chemotherapy and at disease progression.

Data collection and follow up
Study drug administration occurs on Day 1 (± 3 days) of each cycle. Each cycle is 14 days. Cycle 1 should occur within 3 days from registration of pts. All procedures during the study treatment must occur within 3 days prior to the administration, except for radiological assessment required for baseline within 28 days prior to initiation of the study treatment. The following assessments will be performed prior to each cycle every 2 weeks. All radiological assessments will be performed each 8 weeks (± 1 week), regardless of the treatment cycle. CEA will be testing every 8 weeks with radiological assessments. The end of the study treatment visit should occur within 30 days after last dose of study treatment is administered. The post-treatment follow-up visits will occur every 3 months (± 14 days) for 3 years (Table 4).

Statistical analysis and sample size
The primary objective of this study is to assess the ORR, defined as the best response recorded on the ITT population according to RECIST v1.1. In the TRIBE study, ORR for pts. RAS/BRAF mutated treated in first line with FOL-FOXIRI and bevacizumab regimen was 66% [11]. Our hypothesis is that FOLFOXIRI and bevacizumab regimen plus nivolumab is able to improve the ORR from 66 to 80%. An ORR of 80% is considered enough valuable to pursue this combination in a phase III trial. The sample size was calculated using the A'Hern [19] modification of the original Fleming [20] one-stage design. Calculations were performed by the use of PASS Professional v.11.0.10 software [21].
The study requires 64 subjects to decide whether the proportion responding, P, is less than or equal to 0,66 or greater than or equal to 0,80. If the number of responses is 49 or more, the Hypothesis that P < 0,66 is rejected with a target error rate of 0,05 and an actual error rate of 0,046. If the number of responses is 48 or less, the hypothesis that P > 0,800 is rejected with a target error rate of 0,200 and an actual error rate of 0,197. A total 70 pts. will to be enrolled assuming 10% pts. discontinuation rate due to non-compliance or toxicity.

Preliminary safety evaluation
An Independent Monitoring Committee (IDMC) will review safety data 28 days after the inclusion of the 10th patient. Safety data, including demographics, adverse events, serious adverse events, and relevant laboratory data, will be reviewed.
The IDMC will provide a recommendation as to whether the study may continue, whether amendment(s) to the protocol should be implemented, or whether the study should be stopped. The final decision will rest with the Sponsor.

Coordination
Azienda Unità Sanitaria Locale di Reggio Emilia -IRCCS is responsible for the coordination and management of the study on behalf of Gruppo Oncologico Italiano Ricerca Clinica (G.O.I.R.C.) Cooperative Group.

Discussion
The binding of PD-L1 to PD-1 plays a central role in Tcell tolerance by hindering naive and effector T-cell responses. Clinical experience with checkpoint inhibitors has shown that tumors co-opt the PD-L1/PD-1 signaling pathway as one key mechanism to escape immune damage. Nivolumab, an anti-PD-1 monoclonal antibody may block tumor growth in different ways by targeting certain cells.
It's well known that chemotherapy makes the cancer more immunogenic, and more suitable for immunotherapy. Moreover, angiogenetic inhibitors could promote enhanced tumor T-cell infiltration causing in a reprogramming of the tumor microenvironment from immune-suppressive to immune-permissive status. Novel anti-PD-L1 drugs reinforce the action of the antiangiogenetic drugs when administered in combination. Encouraging early indicators of efficacy have been detected with combination strategies using immunecheckpoint inhibitors and biological targeted therapies, such as axitinib in combination with pembrolizumab [46], and nivolumab in combination with sunitinib or pazopanib [47] in mRCC. In another phase 1b study in mRCC, investigating the combination of bevacizumab and an anti -PD-L1, atezolizumab, increased intratumoral CD8 + T-cells and macrophages compared to bevacizumab alone, leading to an increase of MHC I expression, as well as Th-1 and T effector gene signatures in post treatment biopsies assessment [48]. Atezolizumab plus bevacizumab were examined in phase I, II and III studies. The safety of this combination resulted acceptable and AEs leading to treatment interruption were very low. In a phase III study, 40% of pts. treated with atezolizumab plus bevacizumab and 54% of pts. with sunitinib had grade 3-4 AEs; 12 and 8% of allgrade AEs led to discontinuation of treatment, respectively [48,58,59].
Recent findings for enhancement in PFS using bevacizumab and atezolizumab in combination with carboplatin/paclitaxel in front-line lung cancer is a promising strategy, indorsing clinically meaningful and durable benefit for patients [49][50][51].
In a clinical trial conducted in melanoma pts., has been explored the combination of bevacizumab with anti-CTLA-4 inhibitor, ipilimumab, revealed widespread morphological modifications in CD31 + endothelial cells and an extensive tumor penetration of immune cells post-treatment including CD8 + cells and CD163 + macrophages in comparison to ipilimumab treatment alone, thus demonstrating that the combination of anti-VEGF and anti-CTLA-4 inhibitors has the ability to promote immune cell access in the TME [60].
A recent phase III study revealed that in pts. with NSCLC, atezolizumab in addition to bevacizumab plus carboplatin and paclitaxel (ABCP) in 692 pts. with advanced non-squamous NSCLC improve OS (19.2 months vs. 14.7 months; HR 0.78; 95% CI, 0.64 to 0.96; p = 0.02) [51]. The safety profile of ABCP was consistent with safety profiles of each drugs and AEs occurred in 94.4% vs. 95.4% in ABCP and BCP control group, respectively. The most common grade 3 or 4 AEs were febrile neutropenia, and hypertension, and related serious AEs were noticed in 25.4 and 19.3% in the ABCP and BCP groups, respectively. The immune-related AEs (irAEs) grade 1 or 2 occurred in 77.4% of the ABCP group, and the treatment-related deaths occurred in 2.8% of the ABCP group [51].
In mCRC, a phase Ib study examined the safety and efficacy of atezolizumab plus bevacizumab (Arm A) with the dosage of atezolizumab 20 mg/kg q3w and bevacizumab 15 mg/kg q3w versus atezolizumab plus bevacizumab and mFOLFOX6 (Arm B) with atezolizumab 14 mg/kg q2w, bevacizumab 10 mg/kg q2w, and mFOLFOX6 at standard doses. The safety profile in Arm A showed a 64% of grade 3-4 AEs, while in Arm B, 73% pts. had grade 3-4 AEs, especially hematological toxicity. The irAEs grade 3 and 4 were 7 and 20%, respectively. The authors concluded that the addition of atezolizumab plus bevacizumab with or without FOLFOX was well tolerated without unexpected toxicities [61]. Efficacy data are not yet available.
In a phase II study, in 30 mCRC pts., pembrolizumab combined with mFOLFOX6 in first line treatment showed an acceptable toxicity thought suggesting a trend towards an increase of neutropenia; in the initial cohort grade 3 and 4 neutropenia was described but after dose reduction of mFOLFOX6, rate of grade 3 and 4 toxicity was 36.7 and 13.2% with FOLFOX/pembrolizumab and pembrolizumab alone respectively. Best response was partial response in 15 pts. with 100% of disease control rate (DCR) at 8 weeks. After 2 months of therapy, one patient with MMRd had surgical resection accounting complete pathological response. Moreover, the mPFS has not been reached [62].
In conclusion, we assume that there are sufficient evidences to support the combination of treatments with triplet chemotherapy (FOLFOXIRI), antibody anti-VEGF (bevacizumab), and immunotherapy (nivolumab, anti PD-1 antibody) in pts. with mCRC RAS/BRAF mutated, regardless to MMR status.