Effect of vorinostat (VOR) and hydroxychloroquine (HCQ) on immunity and autophagy in metastatic colorectal cancer (mCRC).
670 Background:HCQ enhances the anti-cancer activity of the histone deacetylase inhibitor, VOR in pre-clinical CRC models. Our phase 1 dose escalation trial found that patients (pts) with mCRC obtained prolonged clinical benefit from VOR + HCQ. Thus, a single arm expansion cohort in refractory mCRC was done. Methods: Pts with refractory mCRC received VOR 600 mg by mouth (PO) + HCQ 400 mg PO daily, in a 3-week cycle. The primary endpoint was median progression-free survival (mPFS). Secondary endpoints: median overall survival (mOS), adverse events (AE) (NCI-CTCAEv3.0), flow cytometry (FACS) of peripheral blood mononuclear cells (PBMCs), tumor biopsies. Results: 20 pts were enrolled (19 evaluable for survival): mean age 61 (44-74). Female 7/male 23, 9 Caucasian/10 Hispanic/1 Black, 11 KRAS mutated. ECOG 0-1 (18 pts). 3+ prior lines (13), previous regorafenib (4). Dose level reduction on study (7). mPFS 2.8 months (95% CI: 1.63-8.16). mOS 6.7 months (95% CI: 4.63-NR). Treatment-related grade 3 AEs: nausea/vomiting (3), anemia (3). Grade 4 thrombocytopenia (3). Grade 4 INR elevation (1 pt on Coumadin). No grade 5 AEs. Five pts (26%) had stable disease ≥12 weeks. Treatment significantly reduced regulatory and PD-1+ (exhausted) CD4+ and CD8+ T cells and increased CD45RO+CD62L-CD4+ (effector) T cells, consistent with improved anti-tumor immunity. On-study biopsies (3) showed increased lysosomal protease CTSD and LC3-II consistent with autophagy inhibition. Conclusions: VOR/HCQ is active, safe and tolerated in refractory CRC patients, resulting in potentially improved anti-tumor immunity (decreased exhausted and regulatory T cells and increased effector phenotype T cells) and reduced tumor autophagy. A randomized phase II trial of VOR/HCQ versus regorafenib is now open. Clinical trial information: NCT01023737.