This study will evaluate efficacy and safety of hydroxychloroquine combined with methotrexate, capecitabine and bevacizumab versus regorafenib in participants with refractory metastatic colorectal cancer with mutations in KRAS or NRAS genes. The hypotheses of this study are that a combination of hydroxychloroquine, methotrexate, capecitabine, and bevacizumab (compared to regorafenib) prolongs progression-free survival and overall survival, and also increases rates of objective responses and disease control.

One potential therapeutic option for pretreated patients with RAS-positive metastatic colorectal cancer (mCRC) could be hydroxychloroquine. The KRAS mutation drives uncontrolled proliferation and cell survival through pathways such as MAPK/ERK and PI3K/AKT. The pathways in KRAS-mutant tumor cells activate autophagy to recycle cellular components and sustain growth under stress conditions, such as hypoxia or nutrient deprivation. Autophagy serves as a pro-survival mechanism, enabling tumor cells to resist intrinsic stressors and treatment, such as chemotherapy or radiation therapy. Hydroxychloroquine inhibits autophagy by blocking the fusion of autophagosomes with lysosomes. This results in the accumulation of damaged organelles and toxic metabolites within the cell, ultimately triggering apoptosis. Suppressing autophagy may also disrupt tumor cell metabolism by increasing oxidative stress. In experimental models, KRAS-mutant tumors demonstrated enhanced sensitivity to autophagy inhibitors such as hydroxychloroquine. In murine studies, combining hydroxychloroquine with chemotherapy or MAPK pathway inhibitors, such as trametinib, augmented antitumor efficacy. Multiple studies have reported that hydroxychloroquine achieves rapid objective responses when combined with targeted therapy or chemotherapy. However, rapid development of acquired resistance remains a challenge.

Methotrexate and capecitabine may be considered as options for overcoming acquired resistance to hydroxychloroquine. KRAS-mutant tumors frequently exhibit defects in DNA repair systems. Methotrexate inhibits thymidylate synthesis, causing accumulation of DNA damage, especially uracil misincorporation, which leads to "thymidylate-induced stress" and cell death. Cells with mutations in the KRAS gene may have increased expression of folate transporters, such as RFC or FPGS, making them more sensetive to antifolates such as methotrexate. This increases intracellular drug accumulation and cytotoxic effects. Preclinical studies have shown that exposure to methotrexate in tumor cells with KRAS mutations significantly decreases mRNA expression of the KRAS gene and total levels of KRAS protein. When combined with fluoropyrimidines like capecitabine, methotrexate may enhance suppression of DNA synthesis, which is a critical vulnerability in KRAS-mutant tumors.

Capecitabine increases the sensitivity of KRAS-positive tumors to TRAIL-induced apoptosis, a process that may be amplified by hydroxychloroquine exposure. Bevacizumab can be used to improve the delivery of chemotherapy drugs to tumor cells by increasing drug concentration within the tumor through lowering vascular permeability.

Inclusion Criteria:

• Eastern Cooperative Oncology Group (ECOG) performance status of 0-1
• Provide written informed consent
• Age ≥ 18 years
• Histologically confirmed diagnosis of colorectal cancer (CRC) with distant metastases.
• Presence of mutations in the KRAS or NRAS gene.
• Participants must have previously treated for metastatic colorectal cancer and experienced disease progression during receiving at least 2 lines of systemic chemotherapy in combination with antiangiogenic agents.
• Patient has previously received oxaliplatin- and irinotecan-containing regimens and developed resistance to these chemotherapeutic agents.

Exclusion Criteria:

• Presence of clinically significant cardiovascular disease: severe or unstable ischemic heart disease, history of myocardial infarction, New York Heart Association Class III/IV congestive heart failure, ventricular arrhythmias.
• Stroke and/or transient ischemic attack within 6 months prior to screening;
• Uncontrolled hypertension
• History of previous malignancies except non-melanoma skin cancers, or in situ cervical or breast cancer unless a complete remission was achieved at least 2 years prior to randomization AND no additional therapy is required during the study period. Patients having hepatic involvement of cancer should be excluded as per investigator assessment.
• Patients with CNS metastases are eligible only if the metastases are adequately treated.
• Absolute neutrophil count (ANC) <1.5×109/L, platelet count <100×109/L, or hemoglobin <9.0 g/dL.
• Serum total bilirubin >1.5 × the upper limit of normal (ULN). Participants with Gilbert syndrome, bilirubin <2 X ULN, and normal AST/ALT are eligible;
• Alanine aminotransferase (ALT) or aspartate aminotransferase (AST) >3 × ULN;
• Serum creatinine >1.5 × ULN.
• History of a thromboembolic event
• Presence of any allergic reactions to components of the study drugs
• Concomitant medications with a known risk of causing QT prolongation and/or Torsades de Pointes.
• Any anti-cancer systemic therapy, radiation therapy
• Women who are pregnant or lactating;
• Presence of unresolved adverse events of grade 2 or higher toxicity, according to CTCAE v5.0 criteria, from prior therapy (except for alopecia or neurotoxicity grade≤2).
• Any other serious or uncontrolled medical disorder, active infection, physical examination finding, laboratory finding, altered mental status, or psychiatric condition that, in the opinion of the investigator, would limit a patient's ability to comply with the study requirements, substantially increase risk to the patient, or impact the interpretability of study results.