Therapy with metronomic cyclophosphamide (mCyc) in metastatic castrate resistant prostate cancer (mCRPC) and genomic correlates.

Abstract

e17041 Background: Multiple historic studies demonstrated activity of mCyc in various solid tumors, including mCRPC. Low doses of oral mCyc can lead to enhanced immune stimulation by depleting regulatory T-cells. Its adverse effect profile is well established, with grade 3 and 4 toxicities being rare. In recent years, use of mCyc in mCRPC has fallen out of favor with development of novel therapies. Methods: We conducted a retrospective study of all patients with mCRPC, who received therapy with mCyc for at least 4 weeks at the University of California Irvine and Thomas Jefferson University between 6/1/17 and 1/31/23. Different oral mCyc regimens were included - 50 mg daily continuously, 50 mg daily on days 1 to 14 out of a 28-day cycle, and 250 mg daily on days 1 to 14 out of a 28-day cycle. Results: A total of 20 patients were included, from 62 to 92 years in age. At prostate cancer (PCa) diagnosis, 60.0% (12/20) had de novo metastatic disease. Duration of mCyc therapy was 4 to 41 weeks. Median number of unique PCa therapies received prior to mCyc was 4 [range: 1 to 9], with 75.0% (15/20) pretreated with taxane-based chemotherapy. 25.0% (5/20) of patients achieved a 50% drop in PSA (PSA50), while 30.0% (6/20) achieved PSA reduction but less than PSA50. 10.0% (2/20) had PSA stability without reduction, but experienced improvement in cancer-related symptoms. In those with PSA50, median duration of therapy was 16.3 weeks [range: 14 to 41]. In PSA responders, median time to PSA nadir was 7.2 weeks [range: 1 to 28]. After starting mCyc, 38.5% (5/13) of patients with any PSA or clinical response (i.e. responders) had a transient PSA rise followed by PSA decline, which is suggestive of tumor flare. In responders, 15.4% (2/13) had liver metastasis and 7.7% (1/13) had bone marrow involvement with associated cytopenias. In responders with previous prostate biopsies, 66.6% (6/9) had Gleason 9 or 10 disease. 53.8% (7/13) of responders had genomic data available with a liquid biopsy or tissue sequencing, revealing 3 patients with TP53 mutations and 1 patient with BRCA2 and RB1 co-mutations. 15.0% (3/20) of patients required dose reduction and 20.0% (4/20) dose interruption due to toxicity, with no treatment discontinuations observed. Adverse reactions leading to dose modifications included cytopenias, anorexia, fatigue, nausea, and diarrhea. Conclusions: Despite availability of novel therapeutics for mCRPC, treatment options remain limited. Recognizing the small sample size of this retrospective dataset, oral mCyc showed promising activity with a quick onset of action and a tolerable toxicity profile in heavily pretreated mCRPC patients with aggressive disease features. Given its unique mechanism of action, this provides rationale for future clinical trials, including combining mCyc with other immunomodulating agents such as PD1/PD-L1 inhibitors.