Baseline Circulating Tumor Cell Count as a Prognostic Marker of PSA Response and Disease Progression in Metastatic Castrate-Sensitive Prostate Cancer (SWOG S1216)

Abstract

Purpose: In metastatic castrate-sensitive prostate cancer (mCSPC), combined androgen axis inhibition is a standard of care. Noninvasive biomarkers that guide initial therapy decisions are needed. We hypothesized that CellSearch circulating tumor cell (CTC) count, an FDA-cleared assay in metastatic castrate-resistant prostate cancer (mCRPC), is a relevant biomarker in mCSPC.

Experimental design: SWOG S1216 is a phase III prospective randomized trial of androgen deprivation therapy (ADT) combined with orteronel or bicalutamide for mCSPC. CellSearch CTC count was measured at registration (baseline). Prespecified CTC cut-off points of 0, 1-4, and ≥5 were correlated with baseline patient characteristics and, in a stratified subsample, were also correlated with two prespecified trial secondary endpoints: 7-month PSA ≤0.2 ng/mL versus 0.2-4.0 versus >4.0 (intermediate endpoint for overall survival); and progression-free survival (PFS) ≤ versus >2 years.

Results: A total of 523 patients submitted baseline samples, and CTCs were detected (median 3) in 33%. Adjusting for two trial stratification factors (disease burden and timing of ADT initiation), men with undetectable CTCs had nearly nine times the odds of attaining 7-month PSA ≤ 0.2 versus > 4.0 [OR 8.8, 95% confidence interval (CI), 2.7-28.6, P < 0.001, N = 264] and four times the odds of achieving > 2 years PFS (OR 4.0, 95% CI, 1.9-8.5, P < 0.001, N = 336) compared with men with baseline CTCs ≥5.

Conclusions: Baseline CTC count in mCSPC is highly prognostic of 7-month PSA and 2-year PFS after adjusting for disease burden and discriminates men who are likely to experience poor survival outcomes.

Figure 1.

Consort diagram for baseline CTC analysis in S1216.

Figure 2.

Association of primary, prespecified baseline CTC count categories with 7-month PSA response. Odds ratios (ORs) for achieving partial response (PR, left) or complete response (CR, right) comparing men with ≥ 5 CTCs to 0 CTCs and men with 1–4 CTCs to 0 CTCs. [Sample size for analysis: N=264; 61% CTC 0, 24% CTC 1–4, and 15% CTC 5+].

Figure 3.

Association of secondary baseline CTC count categories with 7-month PSA response using cutpoints previously validated in mCRPC. Odds ratios (ORs) comparing men with ≥5 CTCs to <5 CTCs (A) or men with ≥1 CTCs to 0 CTCs (B) for achieving partial response (PR) or complete response (CR). [Sample size for analysis: N=264; A. 85% CTC 0–4, 15% CTC 5+; B. 61% CTC 0, 39% CTC 1+].

Figure 4.

Association of primary, prespecified baseline CTC count categories with 2-year PFS. Percent of patients who progressed or died at 2 years among men with ≥ 5, 1–4, or 0 CTCs at baseline, with associated ORs. [Sample size for analysis: N=336; 62% undetectable CTC, 24% CTC=1–4, 14% CTC 5+].

Figure 5.

Association of secondary baseline CTC count categories with 2-year PFS using cutpoints previously validated in mCRPC. Percent of patients who progressed or died at 2 years among men with ≥ 5 vs. < 5 (A) or ≥ 1 vs. 0 (B) CTCs at baseline, with associated ORs. [Sample size for analysis: N=336; A. 86% CTC 0–4, 14% CTC 5+; B. 63% CTC 0, 37% CTC 1+].