Abstract P2-11-16: Prospective comparison of prognostic factors in patients starting a third line of chemotherapy for metastatic breast cancer: An ancillary study to the CirCe01 trial

Abstract

Background: Several prognostic factors and composite scores based on either patients general condition and/or usual blood analyses have been described in the setting of advanced metastatic cancers to estimate patients overall survival (OS). Over the past decade, circulating tumor cells (CTC) have been repeatedly reported as being a strong prognostic tool in metastatic breast cancer patients. We took the opportunity of the observational part of a prospective clinical trial (CirCe01, NCT01349842) that included 3rd line metastatic breast cancer patients to assess and compare several prognostic factors/scores and circulating tumor cell count. Patients and Methods: Metastatic breast cancer patients were included before the start of a 3rd line chemotherapy at the Institut Curie, Paris, France. The following adverse prognostic factors were assessed prospectively and compared to overall survival: altered performance status (PS≥2), lymphopenia (<1,000 lymphocyte/microl), elevated CTC count (CellSearch, ≥5CTC/7.5ml), low albumin (<35g/l), elevated CA 15.3 (>30UI/ml), elevated CEA (>5ng/ml), elevated LDH (>50UI/l). Two composite scores were also evaluated: the Barbot score (combining Karnofsky index, number of metastases, albumin and LDH; Barbot JCO 2008) and the Prognostic Inflammatory and Nutritional Index (PINI; combining albumin, prealbumin, orosomucoid and C-reactive protein). Metastatic sites (liver, lung, bone) were also analyzed. Survival was analyzed by Kaplan-Meier curves; multivariate analysis was done using a Cox model. Results: 56 patients have included prospectively and 36 of them (64%) died. CA 15.3 and CEA levels, lymphopenia and metastatic sites had no significant impact on OS in univariate analysis. The incidence of the significant prognostic markers (%), their correlations (p value) and impact (p value) on overall survival are shown. Correlation and impact on overall survival of prognostic markersFactorIncidenceCorrel. with PS (p value)Correl. with Alb (p value)Correl. with LDH (p value)Correl. with Barbot (p value)Correl. with PINI (p value)Univ. OS (p value)CTC > = 545%NSNS<0.001NSNS0.004PS > = 214%-0.0020.04<0.0010.003<0.001Alb <3516%--NS<0.001<0.001<0.001LDH>50065%---0.05NS0.032Barbot score >316%----0.005<0.001PINI >1013%-----<0.001 In multivariable analysis, the three independent prognostic markers on OS were: elevated CTC count (p = 0.003, HR = 3.4 95% CI [1.5-7.7]), poor PS (p = 0.005, HR = 4.1 95% IC[1.5-11.3]) and low albumin (p<0.0001, HR = 11.3 95% IC[3.9-32.5]). Discussion: CTC are an independent prognostic marker, even in advanced metastatic breast cancer. The disease aggressiveness being captured by the CTC count, usual clinical and biochemical tests appear to be sufficient to evaluate the patient prognosis, whereas more complex score (PINI, Barbot) appear to be of low interest. Financed by: Ligue contre le cancer and PHRC 2009. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-11-16.

The clinical utility of molecular diagnostics in breast cancer management

“…In addition, especially in research on liquid biopsies, thresholds of biomarker expression are being used to classify whether the particular biomarker expressions are either present or absent. Assuming that these tools are based on the best available evidence, they may be suitable to facilitate the prioritization of the clinical research agenda when considering different thresholds of the expression a particular biomarker to distinguish different patient subgroups [48].…”

“…Several studies have reported on the discordance in ER/PR/HER2 between tumor needle biopsies and tumor resection. However, most (recent) studies reported on relatively small sample sizes in total, small subgroups of breast cancer patients, or included only patients diagnosed at a single hospital [42,[48][49][50]. In the study presented here, we have evaluated the discordance in ER/PR/HER2 between tumor needle biopsy and tumor resection material in the majority of invasive breast cancer patients diagnosed in 2016 and 2017 in the Netherlands.…”

“…Screening and early detection of cancer have been suggested, e.g. by detecting hypermethylated DNA in urine or blood(48) or circulating proteins and cell-free DNA(6). Unfortunately, clinical research on LBs for screening is scarce and there is not a lot of evidence for use in screening to date.…”