ValproIc Acid to Potentiate Anti-EGFR Treatment Efficacy and Prevent/Revert Resistance in Colorectal Cancer

This study is a multicentric open label academic randomized phase-2 study. The study population will include patients with advanced or metastatic colorectal cancer, RAS/BRAF wt mCRC, eligible for a rechallenge setting (third or later line), with RAS/BRAF wt ctDNA status at study entry. A total of 130 patients (65/arm) will be required. The VICTORIA - Study Part 1 is a phase 2, open label, randozimed study. Before starting study Part 1 treatment, at the time of enrollment, patients will be randomized electronically 1:1 to one of the two arms: ARM A (standard treatment: irinotecan + panitimumab) and ARM B (VPA + irinotecan + panitumumab) Each cycle will be administered every two weeks for both arms. Patients will continue to receive study treatment until treatment failure as previous defined, unacceptable toxicity, physician's decision, patient's refusal, or any other discontinuation criteria. All subjects who finish treatment, whichever the reason, will enter in the follow-up. All patients will be followed until death and data on subsequent treatment will be collected. VPA will be administered per os daily with a titration strategy (table 2 of the protocol) to improve the compliance for the treatment, with a target serum level between 50 and 100 μg/mL that represents the recommended values for epilepsy and also a useful concentration to produce the desired synergistic effect with chemotherapy based on preclinical studies (0.5- 1mM). An initial safety run-in phase involves a safety evaluation after the first 6 patients are randomly assigned to Arm B. Enrollment will be paused after the first 6 patients randomly assigned to Arm B complete 2 cycles of treatment. The enrollment will be interrupted if the treatment combination will be judged not feasible and major safety concerns will arise. A safety review will be conducted, and recommendations will be made by the Steering Committee. The study will continue if there are 2 or fewer adverse events (AEs) of grade 3 or higher deemed related to the addition of valproic acid to the panitumumab + irinotecan. All measurable and non-measurable lesions must be documented at screening (within 28 days prior to randomization) and re-assessed at each subsequent tumor evaluation (every 8 weeks). Tumor assessment by CT scan (chest, abdomen and pelvis) or MRI (abdomen and pelvis); CEA, CA 19.9; and any other tests resulted positive during baseline staging, will be performed at week 8 and every 8 weeks during treatment until treatment failure in both arms. Patients discontinuing study treatment without progressive disease, will undergo tumor assessments every 8 weeks until progressive disease or study withdrawal. Toxicities will be evaluated at each clinical visit throughout the study treatment and up to 4 weeks after last cycle of treatment accordingly to the Common Terminology Criteria for Adverse Events (AEs) of the National Cancer Institute (CTCAE-NCI) version 5.0. Quality of Life will be assessed by the EORTC QLQ-C30 v.3.0 and QLQ-CR29 questionnaire that will be completed by patients at baseline (prior to treatment, once eligibility is confirmed) and every 8 weeks until disease progression, treatment failure or death. At the same time points will be administered selected items of PRO (Patients Reported Outcome) -CTCAE questionnaire. Blood samples will be collected at baseline, during treatment, and at progression. Biomarkers will be correlated with clinical response, patient outcome and toxicity. In addition, biomarkers will be evaluated on tumor tissues from primary tumors or metastases at baseline, when available. To detect a difference between the two groups of 20%; considering 5% of drop out, 61 patients in each arm will be required to achieve 80% power. A null hypothesis of 0.50 is based on the assumption that the median PFS for RAS wt patients rechallenged with anti-EGFR is approximately 16 weeks. An alternative hypothesis of 0.70 represents a potential target of interest for further studies in this setting with the experimental treatment. The test statistic used will be the one-sided stratified (for the stratification factors) Fisher's Exact Test. The significance level of the test is targeted at 0.10. The date of end of study will be the date of the last visit of the last patient. Randomization will be performed with a stratified procedure that will account for tumor sidedness (left vs right), number of metastatic sites (1 vs ≥ 2), previous lines of treatment (2 vs ≥ 3 lines). The VICTORIA - Study Part 2 is a phase 2, single arm study, designed according to the Fleming single-stage design. In patients in ARM A, at time of disease progression, a second study will explore if the addition of VPA to irinotecan and panitumumab might revert the occurrence of resistance. According to the Fleming single-stage design, under the null hypothesis of 0.05, based on the assumption that treatment with standard rechallenge treatment beyond progression usually leads to an extremely low PFS, and under the alternative hypothesis of a PFS rate at 8 weeks of 0.20 with the experimental treatment (a potential target of interest for further studies), considering one-sided alpha and beta errors of 0.05 and 0.20 (80.32) respectively, a total of 27 patients will be required. A null hypothesis of 0.05 is based on the assumption that treatment beyond PD usually leads to an extremely low PFS. An alternative hypothesis of 0.20 represents a potential target of interest for further studies in this setting with the experimental treatment. Null hypothesis will be rejected if at least 4 patients will be free of progression at 8 weeks with the experimental treatment.