Research Paper Volume 14, Issue 8 pp 3464—3483

Improving ovarian cancer treatment decision using a novel risk predictive tool


Figure 4. Determine the prognostic group of 2005 ovarian cancer patients based on TMEscore in GEO and evaluate the predictive ability. (A) K–M curve for OS of different TMEscore groups (log-rank test, P < 0.001). (B, C) According to chemotherapy outcome-stratified analysis (158 ovarian cancer patients), K–M curves in patients with complete response (CR) or non-complete response (non-CR) in different TMEscore group (log-rank test, P = 0.008; log-rank test, P = 0.11). (D) Expression profile of DEGs with survival significance. TMEscore, age, stage, grade, therapy outcome and histology are shown as patient annotations. Top legend, gray indicates missing value. (EG) Forest plots illustrate the results of multivariate Cox proportional hazards model of clinical feature in all patients, CR patients and non-CR patients respectively.