Dynamic classification using case‐specific training cohorts outperforms static gene expression signatures in breast cancer. Issue 9 (11th October 2014)

• Record Type: Journal Article
• Authors: Győrffy, Balázs; Karn, Thomas; Sztupinszki, Zsófia; Weltz, Boglárka; Müller, Volkmar; Pusztai, Lajos
• Is Part Of: International journal of cancer. Volume 136:Issue 9(2015:May 01)
• Page Start: 2091
• Page End: 2098
• Publication Date: 2014-10-11
• Subjects: Cancer -- Periodicals; Cancer -- Prevention -- Periodicals; 616.994
• DOI: 10.1002/ijc.29247
• Languages: English
• ISSNs: 0020-7136

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <p>The molecular diversity of breast cancer makes it impossible to identify prognostic markers that are applicable to all breast cancers. To overcome limitations of previous multigene prognostic classifiers, we propose a new dynamic predictor: instead of using a single universal training cohort and an identical list of informative genes to predict the prognosis of new cases, a case‐specific predictor is developed for each test case. Gene expression data from 3, 534 breast cancers with clinical annotation including relapse‐free survival is analyzed. For each test case, we select a case‐specific training subset including only molecularly similar cases and a case‐specific predictor is generated. This method yields different training sets and different predictors for each new patient. The model performance was assessed in leave‐one‐out validation and also in 325 independent cases. Prognostic discrimination was high for all cases (<italic>n</italic> = 3, 534, HR = 3.68, <italic>p</italic> = 1.67 <italic>E−</italic>56). The dynamic predictor showed higher overall accuracy (0.68) than genomic surrogates for Oncotype DX (0.64), Genomic Grade Index (0.61) or MammaPrint (0.47). The dynamic predictor was also effective in triple‐negative cancers (<italic>n</italic> = 427, HR = 3.08, <italic>p</italic> = 0.0093) where the above classifiers all failed. Validation in independent patients yielded similar classification power (HR = 3.57). The dynamic classifier is available online at <ext-link ext-link-type="uri" xlink:href="http://www.recurrenceonline.com/?q=Re_training" xlink:type="simple" xmlns:xlink="http://www.w3.org/1999/xlink">http://www.recurrenceonline.com/?q=Re_training</ext-link>. In summary, we developed a new method to make personalized prognostic prediction using case‐specific training cohorts. The dynamic predictors outperform static models developed from single historical training cohorts and they also predict well in triple‐negative cancers.</p> </abstract>