ctDNA Positivity Correlates With Poor Survival Outcomes in mCRPC

May 31, 2020
Caroline Seymour

"Liquid biopsies continue to gain attention in the field of oncology. The ability to have a rapid, noninvasive test, and one that allows for serial treatment monitoring is imperative for more precise patient care."

Circulating tumor DNA (ctDNA) showing positivity in patients with metastatic castration-resistant prostate cancer (mCRPC) at baseline and posttreatment in the randomized phase 2 A.MARTIN trial was associated with worse radiologic progression-free survival (rPFS), according to a correlative analysis presented at the 2020 Virtual Scientific Meeting.1

In the trial, 253 patients with mCRPC who had received prior docetaxel-based therapy, progressed on at least 1 hormonal therapy, and had an ECOG performance score of 0 to 1, were randomized to 28-day cycles of 1 of 3 treatment regimens: abiraterone acetate (Zytiga) and placebo, abiraterone and 200 mg of daily ipatasertib, or abiraterone and 400 mg of daily ipatasertib.

The median rPFS was 5.7 months in patients with ctDNA positivity versus 8.3 months in patients with ctDNA negativity at baseline (HR, 1.9; 95% CI, 1.3-2.7; P =.00035). The correlation between ctDNA positivity and worse rPFS was upheld in a multivariate Cox model that accounted for baseline clinical variables, such as age, stage, ECOG performance status (PS), Gleason sum, visceral metastases, prior chemotherapy, measurable disease, prostate-specific antigen (PSA), hemoglobin, alkaline phosphatase, and albumin (HR, 1.2; 95% CI, 1.2-2.9; P <.01).

“Liquid biopsies continue to gain attention in the field of oncology. The ability to have a rapid, noninvasive test, and one that allows for serial treatment monitoring is imperative for more precise patient care,” lead study author, Jane Goodall, BSc, research scientist at The Institute of Cancer Research, said in a virtual presentation during the meeting. “Liquid biopsies hold promise, as they can act as diagnostic, prognostic, and predictive biomarkers in the clinic.”

The primary results of the trial, which were published in Clinical Cancer Research,2 showed that the combination of abiraterone and ipatasertib at 400 mg led to a significant improvement in rPFS in patients with PTEN loss (HR, 0.39; 95% CI, 0.22-0.70). Specifically, median rPFS in the 400-mg ipatasertib arm was 11.5 months versus 4.5 months in the placebo arm. Patients without PTEN loss had a similar median rPFS of 7.5 months with ipatasertib versus 5.6 months with placebo arm (HR, 0.84; 95% CI, 0.51-1.37).

In the correlative analysis, ctDNA in the plasma was analyzed with a prostate-cancer specific 58-gene panel and sequenced to a depth of 2500X. Nonsynonymous mutations were identified in the plasma, and all germline variants were excluded from analysis. Germline mutations were subtracted using peripheral blood mononuclear cell DNA, and all other mutations that occurred in approximately 50% allele frequency were excluded from the analysis to approximate germline filtering.

“We reported the variant allele frequency of somatic mutations, thereby representing the true circulating tumor fraction at all available timepoints,” said Goodall.

ctDNA was analyzed at baseline (n = 207), at the start of cycle 3 (n = 206), and at the end of treatment (n = 162). In total, 143 patients underwent ctDNA assessment at all 3 timepoints.

ctDNA positivity was defined as more than 1% allele frequency. The most commonly mutated genes that were detected at baseline included somatic TP53 (30%), AR (9%), FOXA1 (8%), PTEN (8%), and SPOP (6%) mutations.

Additional findings revealed a correlation between prognostic factors and ctDNA levels at baseline. The clinical factors that had the strongest correlation with high levels of baseline ctDNA included ECOG PS (P =.00097), PSA (P =.00081), hemoglobin (P =.0025), and alkaline phosphatase (P =.00008).

Patients with an ECOG PS of 1, PSA greater than 250 ng/mL, hemoglobin less than the median, and alkaline phosphatase at or above the median had significantly higher levels of ctDNA at baseline.

According to mutational profile, TP53, FOXA1, and PI3K/AKT pathway gene mutations showed worse rPFS in the abiraterone-alone arm (HR, 1.6; HR, 4.7; HR, 1.2, respectively) versus the 400-mg ipatasertib arm (HR, 1.1; HR, 2.3; HR, 0.89, respectively).

Investigators also evaluated the dynamic changes of ctDNA and found that posttreatment ctDNA from baseline to cycle 3 showed a correlation between the depth of response and ctDNA negativity. Patients who achieved a complete response or partial response had the greatest reduction in the variant allele frequency from baseline to cycle 3 of treatment, with a mean ctDNA reduction of 18.1%.

A similar trend was observed in patients who achieved stable disease and/or PSA response, with a mean reduction in the variant allele frequency of 15.9%. A smaller number of patients who achieved stable disease and no PSA response experienced a reduction in ctDNA levels, although the overall mean reduction was 7.0%. Patients who developed progressive disease experienced a constant variant allele frequency, with a mean ctDNA elevation of 0.82%.

Regarding the association between posttreatment ctDNA status and rPFS, patients who sustained ctDNA negativity from baseline to cycle 3 of treatment experienced the most favorable outcomes, with a median rPFS of 8.3 months (7.6-11.4; HR, 1.55; 95% CI, 1.05-2.29; P =.0279). Patients who had ctDNA at baseline and achieved some level of reduction in ctDNA by cycle 3 of treatment experienced a comparable median rPFS of 8.1 months (5.7-10.3). Higher posttreatment ctDNA was significantly correlated with worse rPFS, leading to a median rPFS of 4 months (2.8-5.6).

The correlation was also apparent among patients with an increase in ctDNA versus sustained ctDNA negativity (HR, 2.89; P <.0001), patients with an increase in ctDNA versus a decrease in ctDNA (HR, 2.16; P = .0008), and patients with a decrease in ctDNA versus sustained ctDNA negativity (HR, 1.55; P =. 0279).

Further validation of these data are anticipated from the ongoing phase 3 IPATential150 trial (NCT03072238), in which patients with mCRPC randomized to ipatasertib and prednisone versus placebo plus abiraterone and prednisone will undergo serial plasma assessment.

“ctDNA analyses may help identify poorer prognosis disease at baseline, inform on treatment response and rPFS in on-treatment samples, and elucidate emerging resistance mechanisms at disease progression,” Goodall concluded.


1. Goodall J, Assaf ZJ, Shi Z, et al. Circulating tumor DNA (ctDNA) dynamics associate with treatment response and radiological progression-free survival (rPFS): analyses from a randomized phase II trial in metastatic castration-resistant prostate cancer (mCRPC). J Clin Oncol. 2020;38(suppl 15): 5508. doi:10.1200/JCO.2020.38.15_suppl.5508

2. S. de Bono J, De Giorgi U, Nava Rodrigues D, et al. Randomized phase II study evaluating Akt blockade with ipatasertib, in combination with abiraterone, in patients with metastatic prostate cancer with and without PTEN loss. Clin Cancer Res. 2019;25(3): 928-936. doi:10.1158/1078-0432.CCR-18-0981

<< View more of Targeted Oncology's coverage from the 2020 ASCO Virtual Annual Meeting