Changing metastatic patterns associate with dynamics of circulating tumor DNA in metastatic castration-resistant prostate cancer

Abstract

Background: Circulating tumor DNA (ctDNA) acts as an early biomarker of the efficacy of androgen receptor signaling inhibitor (ARSI) therapy. In this study, we aimed to reveal if ctDNA can supplement imaging to better predict metastasis burden and radiographic progression disease (PD) in metastatic castration-resistant prostate cancer (mCRPC).

Methods: Targeted next-generation sequencing was performed to assess ctDNA fraction. Radiographic evidence was documented by conventional imaging according to Prostate Cancer Working Group 3 criteria.

Results: We prospectively collected plasma samples from 112 mCRPC with bone (n = 77), lymph nodal (n = 31), and visceral (n = 4) metastases. Only bone metastatic patterns were significantly associated with median ctDNA at baseline, during treatment and at PD (P <.0001). At first radiographic restaging, 24 (31.2%) men with a progressive worsening of bone disease had early ctDNA rise with a % ctDNA variation of 150.6% (interquartile range [IQR] = 104.9-210.7] compared with 11.1% (IQR = 0-36.6), P <.0001, in men with no change in bone disease. Univariate analysis showed that early ctDNA rise was significantly associated with progression free/overall survival (PFS/OS). In multivariable analysis including ctDNA change from baseline to 3-month treatment, variation of bone metastatic patterns (from oligometastatic to polymetastatic and/or to widespread disease), presence of visceral metastasis, age, PSA, performance status and prior docetaxel therapy, the transition from low- to high-ctDNA within 3 months of starting ARSI therapy was a significant predictor of OS (HR = 2.50, 90% CI, 1.06-5.88, P =.035) and persistent high level of ctDNA was a predictor of PFS (HR = 2.53, 95% CI, 1.10-5.81, P =.028). Metastatic involvement demonstrated that the transition from bone polymetastatic to widespread disease and the presence of visceral metastases were both associated with worse OS (HR = 2.43, 95% CI, 1.10-5.35, P =.028, and HR = 3.40, 95% CI, 1.50-7.66, P =.003, respectively). Prior therapy with docetaxel represented an independent predictor of both PFS and OS (HR = 2.47, 95% CI, 1.40-4.35, P =.002, and HR = 1.78, 95% CI, 1.00-3.15, P =.049, respectively).

Conclusions: Early ctDNA variation might reflect changes in metastatic burden and, likely, in bone metastatic patterns on ARSI therapy allowing to track pattern of disease progression and to predict outcome.

Keywords: androgen receptor signaling inhibitors; bone metastasis; ctDNA; mCRPC; metastatic pattern.

Figure 1.

Association between variation in ctDNA fraction and pattern of bone metastases. (A) and (B) The percentage change in ctDNA fraction and the transition of bone metastatic pattern from baseline to 3-month treatment. (C) Exemplar case of mCRPC patient receiving abiraterone with transition from bone oligometastatic to bone diffuse disease with an increase of ctDNA fraction. (D) Exemplar case of mCRPC man treated with enzalutamide showing a concomitant stable widespread (S) pattern and ctDNA value at baseline and 3-month treatment. Abbreviations. ctDNA, circulating tumor DNA; O, oligometastatic (≤5 lesions); D, polymetastatic (between 6 and 19 lesions); S, widespread (≥20 lesions).

Figure 2.

Association between early ctDNA variation and clinical outcome. (A) Progression-free survival (PFS) and (B) overall survival (OS) according to variation in ctDNA fraction from baseline to 3-month treatment (Low-Low, Low-High, and High-High); (C) Univariate analysis of PFS and OS as a function of ctDNA change.