CTC counts as a biomarker of prognosis and response in metastatic castration-resistant prostate cancer (mCRPC) from the CARD trial.

Abstract

161 Background: In the prospective CARD trial (NCT02485691), cabazitaxel significantly improved radiographic progression-free survival (rPFS) and overall survival (OS) versus abiraterone or enzalutamide in patients with metastatic castration-resistant prostate cancer who had received docetaxel and progressed within 12 months with the alternative androgen-signaling-targeted inhibitor. Here we analyzed circulating tumor cell (CTC) counts as a biomarker of prognosis and response to therapy. Methods: Blood samples collected at screening (SC), Cycle 2 (C2) Day 1, and end of therapy were sent to Epic Sciences for CTC analysis. The association between CTC counts, defined as any cytokeratin-positive, CD45-negative cell, and clinical outcomes were pre-specified and analyzed using a multivariate Cox model for time-to-events (rPFS, OS) or a Χ2 test for categorical outcomes (objective tumor response). Results: Of the 255 patients randomized, 237 (92.9%) had evaluable samples at SC and 213 (83.5%) at C2. CTCs ≥ 1 were detected in 204 samples at SC (86%) and 178 samples at C2 (84%), with a median (interquartile range) count per mL of blood of 4.1 (1.3–14.2) and 5.2 (1.5–17.3), respectively. At baseline, higher CTC counts were associated with higher values of lactate dehydrogenase (P = 0.014), alkaline phosphatase (P < 0.001), and prostate-specific antigen (P < 0.001). High CTC counts, measured as a continuous variable at SC (all arms combined) were associated with a shorter OS (P = 0.03), but not rPFS (P = 0.7). However, favorable changes in absolute CTC count from SC to C2, all arms combined, were associated with longer rPFS (P = 0.03). Conversion from high (≥ 3) to low (< 3) CTC count at C2 or maintenance of a low CTC count at C2 was associated with objective tumor response (P = 0.007). Analysis of the associations of CTC androgen receptor variant 7, chromosomal instability, heterogeneity status (Shannon Index), single-cell genotypes (e.g. retinoblastoma/phosphatase and tensin homolog copy-loss), and the presence of CTCs with small-cell/neuroendocrine morphology is ongoing. Conclusions: This preplanned analysis of the CARD trial confirms that baseline CTC counts, measured by Epic Sciences platform, are prognostic. Moreover, early favorable changes in CTC counts from baseline to C2 are associated with response to therapy. Funding: Sanofi. Clinical trial information: NCT02485691.