Baseline genomic and circulating tumor cell (CTC) correlative data from very high-risk (VHR), localized, node-negative prostate cancer patients.

person Sean Matthew McBride Memorial Sloan Kettering Cancer Center, New York, NY info_outline Sean Matthew McBride, Michael J. Zelefsky, Daniel Eidelberg Spratt, Channing Judith Paller, Marisa Kollmeier, Susan F. Slovin, Jahan Aghalar, Jason W.D. Hearn, Robert Benjamin Den, Curtiland Deville, Han Xiao, Wassim Abida, Howard I. Scher, Dana E. Rathkopf

Authors person Sean Matthew McBride Memorial Sloan Kettering Cancer Center, New York, NY info_outline Sean Matthew McBride, Michael J. Zelefsky, Daniel Eidelberg Spratt, Channing Judith Paller, Marisa Kollmeier, Susan F. Slovin, Jahan Aghalar, Jason W.D. Hearn, Robert Benjamin Den, Curtiland Deville, Han Xiao, Wassim Abida, Howard I. Scher, Dana E. Rathkopf Organizations Memorial Sloan Kettering Cancer Center, New York, NY, Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins University, Baltimore, MD, Memorial Sloan Kettering Cancer Center, Commack, NY, University of Michigan, Ann Arbor, MI, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA, Johns Hopkins University School of Medicine, Baltimore, MD, Memorial Sloan Kettering Cancer Center, Basking Ridge, NJ Abstract Disclosures Research Funding Pharmaceutical/Biotech Company Other Foundation Background: Few data exist on CTC frequency and number, genetic landscape, and tumor mutational burden (TMB) in VHR, node negative, localized prostate cancer (PCa). Methods: We are conducting a single-arm phase 2 trial of ultra-hypofractionated radiation (RT) with 6 months of abiraterone, apalutamide and leuprolide in VHR PCa, defined as: Gleason (Gl) 9-10 or 2 high risk features (including radiographic (r) T3/T4 disease) or > 4 cores of Gl8. We report baseline correlatives in the first 38 screened patients (pts). CTCs were isolated using a non-selection based platform (EPIC Sciences). Additional analyses were conducted using MSK IMPACT, a next generation sequencing assay. Results: Median PSA was 14.8 ng/mL (IQR, 7.7-28.1); Gl7 was present in 5% (n = 2), Gl8 in 32% (n = 12), Gl9 in 61% (n = 23) and Gl10 in 2% (n = 1) of pts; on MR, 42% of pts were rT2 (n = 16), 39% had rT3a disease (n = 15) and 18% had rT3b disease (n = 7). CTC data were available on 31 pts; 74% (n = 23) had ≥1 detectable CTC (range, 0.8-14.6 cells per mL); 29% (n = 9) had CK+ clusters (range, 0.8-7.1 clusters per mL). IMPACT was available for 20 pts: KMT2D/C mutations were present in 25% (n = 4), TP53 missense mutations in 20% (n = 4), FOXA1 mutations in 20% (n = 4), PTEN truncating or missense mutations in 15% (n = 3), SPOP missense mutations in 15% (n = 3), PIK3CA activating mutations in 15% (n = 3), APC deletions in 15% (n = 3); 85% (17/20) had alterations in one of these genes. No clinically significant germline mutations were present. Median TMB was 2.63 mutations/mB (range, 0.87-60.56); the TMB-highest pt had an in-frame deletion in MSH2. Among IMPACTed pts with normalized testosterone post-protocol treatment (n = 16), there was a trend towards an association with SPOP/FOXA1 mutations and undetectable ( < 0.05 ng/mL) PSA; 5/6 pts with mutations had undetectable PSA (83.3%) vs 3/10 without (30%) (p = 0.12). The trial is managed by the PCCTC and funded by Janssen. Conclusions: The genetic profile and TMB of VHR, localized PCa resembles non-castrate, metastatic disease. The frequency of detectable CTCs is high with implications for post-treatment surveillance. SPOP/FOXA1 mutations may predict initial response to RT with total androgen blockade. Clinical trial information: NCT02772588