Exploratory Analysis of TP53 Mutations in Circulating Tumour DNA as Biomarkers of Treatment Response for Patients with Relapsed High-Grade Serous Ovarian Carcinoma: A Retrospective Study

Abstract

Background

Circulating tumour DNA (ctDNA) carrying tumour-specific sequence alterations may provide a minimally invasive means to dynamically assess tumour burden and response to treatment in cancer patients. Somatic TP53 mutations are a defining feature of high-grade serous ovarian carcinoma (HGSOC). We tested whether these mutations could be used as personalised markers to monitor tumour burden and early changes as a predictor of response and time to progression (TTP).

Methods and Findings

We performed a retrospective analysis of serial plasma samples collected during routine clinical visits from 40 patients with HGSOC undergoing heterogeneous standard of care treatment. Patient-specific TP53 assays were developed for 31 unique mutations identified in formalin-fixed paraffin-embedded tumour DNA from these patients. These assays were used to quantify ctDNA in 318 plasma samples using microfluidic digital PCR. The TP53 mutant allele fraction (TP53MAF) was compared to serum CA-125, the current gold-standard response marker for HGSOC in blood, as well as to disease volume on computed tomography scans by volumetric analysis. Changes after one cycle of treatment were compared with TTP.

The median TP53MAF prior to treatment in 51 relapsed treatment courses was 8% (interquartile range [IQR] 1.2%–22%) compared to 0.7% (IQR 0.3%–2.0%) for seven untreated newly diagnosed stage IIIC/IV patients. TP53MAF correlated with volumetric measurements (Pearson r = 0.59, p < 0.001), and this correlation improved when patients with ascites were excluded ( r = 0.82). The ratio of TP53MAF to volume of disease was higher in relapsed patients (0.04% per cm ³) than in untreated patients (0.0008% per cm ³, p = 0.004). In nearly all relapsed patients with disease volume > 32 cm ³, ctDNA was detected at ≥20 amplifiable copies per millilitre of plasma. In 49 treatment courses for relapsed disease, pre-treatment TP53MAF concentration, but not CA-125, was associated with TTP. Response to chemotherapy was seen earlier with ctDNA, with a median time to nadir of 37 d (IQR 28–54) compared with a median time to nadir of 84 d (IQR 42–116) for CA-125. In 32 relapsed treatment courses evaluable for response after one cycle of chemotherapy, a decrease in TP53MAF of >60% was an independent predictor of TTP in multivariable analysis (hazard ratio 0.22, 95% CI 0.07–0.67, p = 0.008). Conversely, a decrease in TP53MAF of ≤60% was associated with poor response and identified cases with TTP < 6 mo with 71% sensitivity (95% CI 42%–92%) and 88% specificity (95% CI 64%–99%). Specificity was improved when patients with recent drainage of ascites were excluded. Ascites drainage led to a reduction of TP53MAF concentration. The limitations of this study include retrospective design, small sample size, and heterogeneity of treatment within the cohort.

Conclusions

In this retrospective study, we demonstrated that ctDNA is correlated with volume of disease at the start of treatment in women with HGSOC and that a decrease of ≤60% in TP53MAF after one cycle of chemotherapy was associated with shorter TTP. These results provide evidence that ctDNA has the potential to be a highly specific early molecular response marker in HGSOC and warrants further investigation in larger cohorts receiving uniform treatment.

Abstract

James Brenton and colleagues reveal tumour burden and prognosis correlations from circulating DNA analysis in high grade serous ovarian carcinoma.

Author Summary

Why Was This Study Done?

• The standard clinical blood test for measuring response in women receiving chemotherapy for high-grade serous ovarian cancer (HGSOC) is the serum protein cancer antigen 125 (CA-125).

• CA-125 is sensitive but it lacks specificity for detection of ovarian cancer, and in response to chemotherapy, CA-125 level does not change rapidly enough to suggest change in treatment after one or two cycles if chemotherapy treatment is ineffective.

• Better tumour markers are required, and circulating tumour DNA (ctDNA) is a promising candidate.

• ctDNA is cell-free DNA derived from tumour cells that can be detected in the bloodstream; ctDNA can be used as a highly specific marker because it carries mutations unique to the cancer.

What Did the Researchers Do and Find?

• HGSOC is an ideal cancer to test ctDNA as a biomarker because 99% of patients have a mutation in the TP53 gene.

• We designed patient-specific TP53 assays for a retrospective study of 40 patients with HGSOC, and these were used to quantify the amount of ctDNA in 318 plasma samples collected before, during, and after chemotherapy.

• We asked if ctDNA level was correlated with the amount of disease present before chemotherapy treatment measured using 3-D volume reconstruction from CT images taken as part of routine care.

• We also asked if the decrease in TP53 ctDNA after one cycle of chemotherapy treatment could predict which patients would have progression of their cancer within six months.

• ctDNA level, but not CA-125 level, was strongly correlated with the total volume of disease.

• Patients whose ctDNA level exhibited a decrease of >60% after one cycle of chemotherapy had a significantly longer time to progression than those whose ctDNA level decreased by 60% or less.

What Do These Findings Mean?

• TP53 ctDNA has the potential to be a clinically useful blood test to assess prognosis and response to treatment in women with HGSOC.

• The response findings from this retrospective study should be confirmed in larger, prospective studies with uniform treatment. If these findings are confirmed, TP53 ctDNA could be used in HGSOC clinical trials and routine practice to identify earlier whether treatment is effective.