Exploration Of Circulating Tumor DNA As A Therapeutic Response And Prognostic Biomarker In Advanced Breast Cancer

[Objectives]TP53 mutations are common in breast cancer.There is currently no large-scale cohort study to investigate the TP53 landscape in breast cancer patients from China.The predictive value of TP53 mutations for the efficacy of human epidermal growth factor receptor 2(HER2)-targeted therapy in breast cancer remains controversial.In the present study,we aimed to analyze the clinical spectrum and prognostic value of TP53 mutations in circulating tumor DNA(ctDNA)from breast cancer patients in China.[Methods]Metastatic breast cancer treated at Cancer Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College(Beijing,China)between March 2015 and October 2018.Ten millilitres of peripheral blood were collected from each patient before receiving anti-cancer therapy.TP53 mutations were detected by next-generation sequencing.We compared the TP53 mutation frequency in our cohort with the MSK-IMPACT cohort.The gene mutation status and clinical information of 1746 patients with breast cancer were downloaded from the MSK-IMPACT dataset.Kaplan-Meier analysis was performed to estimate the survival curves of the different subgroups,and the log-rank test was used to compare the curves.A multivariable Cox regression model was used to estimate TP53 mutation status and clinical characteristics associated with the survival.[Results]A total of 804 female patients with metastatic breast cancer were investigated in the present study.Among the 804 investigated patients,431(53.6%)patients harbored TP53 mutations.The majority of the mutations were missense mutations(55.7%)and 75.1%of the mutations clustering in exons 5-8,mostly spanning the sequence encoding the DNA-binding domain.TP53 mutations were differentially distributed among different molecular subtypes of breast cancer(P<0.001).TP53 mutation frequency was lower in luminal-type breast cancer patients(44.7%)than in HER2+/hormone receptor-patients(69.4%,P<0.001)and triple-negative breast cancer(TNBC)patients(73.0%,P<0.001).Patients with hormone receptor-/HER2+tumors showed a higher proportion of span mutations as compared with non-hormone receptor-/HER2+patients(17.1%versus 8.0%;P=0.005),whereas those with hormone receptor+/HER2+cancer showed a higher proportion of missense mutations compared with patients with non-hormone receptor+/HER2+cancer(67.1%versus 53.7%;P=0.030).A total of 444 patients recorded disease-free survival(DFS).Patients with TP53 mutations had a shorter DFS than those with wild-type TP53(HR=1.32;95%CI,1.09-1.61;P=0.005).In multivariate Cox regression analysis,TP53 mutation status was not significantly associated with DFS(HR=1.1 8;95%CI,0.97-1.45;P=0.101).However,the DFS of patients with TP53 mutations in exons 5-8 was significantly shorter than the DFS of patients with wild-type TP53(HR=1.50;95%CI,1.11-2.03;P=0.009).TP53 mutation status was not significantly associated with progression-free survival(PFS)in HER2-positive patients who received first-line trastuzumab-based therapy(HR=1.01;95%CI,0.59-1.75;P=0.966).Interestingly,in the taxane combination group,patients with TP53 mutations exhibited longer PFS than those without TP53 mutations(HR=0.08;95%CI,0.02-0.30;P<0.001).However,in the non-taxane combination group,patients with TP53 mutations displayed shorter PFS than those with wild-type TP53(HR=4.84;95%CI,1.60-14.66;P=0.005).[Conclusions]TP 5 3 are more frequently observed in HER2+/hormone receptor-breast cancer patients and those with TNBC than in those with other subtypes of breast cancer.TP53 mutations encoding the DNA-binding domain may be an independent prognostic marker for short DFS in patients with metastatic breast cancer.Patients with TP53 mutation are more likely to benefit from trastuzumab-based chemotherapy in combination with taxanes.TP53 mutation is related to the drug-resistance of trastuzumab-based chemotherapy in combination without taxanes.[Objectives]Molecular tumor burden based on circulating tumor DNA(ctDNA)are potential to monitor anti-cancer treatment response.Ongoing challenges to routine use of ctDNA in clinical practice include validation of results in larger patient cohorts,and demonstration of added clinical utility beyond routine radiologic assessment.We conduct this study to explore the role of molecular tumor burden index(mTBI)in ctDNA as a therapeutic response and prognostic biomarker in a larger cohort prospective phase ?randomized multicenter study.[Methods]This is an exploratory analysis of the CAMELLIA study(NCT01917279).Ten millilitres of peripheral blood were collected from patients who consented to participate in biomarker analysis study.Target-capture deep sequencing of 1021 genes was performed to detect somatic variants in ctDNA from plasma samples.The mTBI value of each sample is calculated by the mathematical model constructed in this study.Kaplan-Meier analysis was performed to estimate the survival curves of the different subgroups,and the log-rank test was used to compare the curves.A multivariable Cox regression model was used to estimate mTBI and clinical characteristics associated with the survival and response.A receiver operation characteristic(ROC)analysis was performed to evaluate the ability of mTBI to predict treatment response.[Results]A total of 291 peripheral blood samples from 125 patients were collected for ctDNA analysis.We detected ctDNA in the pretreatment samples of 94(80.3%)out of 117 patients.An undetectable ctDNA at baseline was associated with a lower disease volume,the mean tumor size in the patients who were not detected ctDNA in the pretreatment was 2.8 cm,which was lower than 4.5 cm in the patients detected mutations in the pretreatment(P=0.045).There are 56.8%patients(71/125)were detected at least one mutation related to resistance of some endocrine therapy.And,55.2%patients(69/125)had at least one actionable event for which an FDA-approved drug is currently available.We defined mTBI based on a comprehensive analysis of somatic variations in ctDNA,with consideration heterogeneity and dynamic evolution.The median mTBI value in the pretreatment of all the 117 patients was 2.2(range 0-36.0).The value of mTBI in pretreatment was correlated with tumor burden(P=0.025).Patients with high-level pretreatment mTBI have shorter OS than patients with low-level pretreatment mTBI,the median OS was 40.9 months and 68.4 months,respectively(P=0.011).Patients with mTBI decreased less than 0.02%in the first tumor evaluation have shorter progression free survival(PFS)and overall survival(OS)than patients with mTBI more than 0.02%in first tumor evaluation(median PFS was 7.2 months versus 10.9 months,P<0.001;median OS was 43.5 months versus 68.1 months,P=0.007).In the ROC analysis of mTBI values,the area under the curve was 0.98(P=0.002).The mTBI is good sensitivity that identified the partial response/complete response and progressive disease based on CT scan(88.5%and 87.5%,respectively).However,72.5%of the 91 samples,which classified as stable disease(SD)based on CT scan,identified as mPR based on ctDNA.Patients achieved the molecular response have a longer PFS and OS compared to non-molecular responders(median PFS was 9.7 months versus 6.8 months,P<0.001;median OS was 64.7 months versus 33.5 months,P=0.036),While the CT evaluation did not(P>0.05).Patients classified as the molecular responders have longer PFS and OS than non-molecular responders for the patients who defined as SD based on CT scan(median PFS was 10.2 months versus 7.2 months,P=0.027;median OS was 64.7 months versus 33.5 months,P=0.015).[Conclusions]The pretreatment and early change of mTBI could be a prognostic biomarker for breast cancer patients treated with chemotherapy.mTBI could be potentially used as a response evaluation criterion and could predict the long-term survival and adjudicates which patients with initial radiologically SD will ultimately respond to anti-cancer therapy.