Predicting outcomes in female germ cell patients using a modified International Germ Cell Cancer Collaborative Group classification system to guide management (100)

Abstract

<b>Objectives:</b> The International Germ-Cell Cancer Collaborative Group (IGCCCG) classification guides testicular germ-cell tumor (GCT) management, but there is no validated risk-based model for female patients (pts). We devised modified IGCCCG (mIGCCCG) classifications for female GCT pts undergoing chemotherapy, validated it in a separate cohort, and investigated clinical-pathologic factors associated with survival outcomes. <b>Methods:</b> An mIGCCCG risk classification was created in a test cohort of female GCT pts treated at a single center (1990-2012). In non-dysgerminomas, good, intermediate and poor-risk groups were stratified by tumor markers or metastatic site (AFP<1,000 ng/mL, HCG <5,000 mIU/mL or LDH<1.5x upper limit of normal (ULN); AFP 1,000-10,000 ng/mL, HCG 5,000-50,000 mIU/mL or LDH 1.5-10xULN; AFP>10,000 ng/mL, HCG>50,000 mIU/mL or LDH>10xULN OR non-pulmonary visceral [excluding peritoneal] metastases, respectively). In dysgerminomas, good and intermediate-risk groups were segregated by non-visceral metastases. The mIGCCCG risk model was validated in an independent cohort treated at the same institution (1986-2020). Progression-free survival (PFS) and overall survival (OS) were estimated among all pts using the Kaplan-Meier method. Differences by clinical-demographic factors were assessed with the log-rank test and Cox proportional-hazards models. <b>Results:</b> Median age, race, stage, histology, and chemotherapy regimen were similar between the original (<i>n</i>=93) and validation (<i>n</i>=94) cohorts. As in the original cohort, both preoperative and pre-chemotherapy mIGCCCG models were significantly associated with outcome in the validation and combined cohorts (Figure 1). Using pre-chemotherapy tumor markers, estimated 3-year OS rates for good, intermediate, and poor-risk pts were 89%, 92%, and 45%, respectively, in the original cohort, whereas 100%, 80%, and 50% in the validation cohort and 94%, 88%, and 51% in the combined cohort (all p-values <0.001). Using preoperative tumor markers, estimated 3-year OS rates for good, intermediate, and poor-risk pts were 92%, 86%, and 46%, respectively, in the original cohort, whereas 100%, 86%, and 60% in the validation cohort, and 96%, 86% and 54% in the combined cohort (all p-values <0.01). On multivariable regression analyses, worse PFS outcomes were associated with the higher stage (HR: 4.3, 95% CI: 1.5-12.0, p=0.006), non-dysgerminoma histology (HR: 3.4, 95% CI: 1.8-6.4, p<0.001), and older age (HR: 1.17 per 5-year increase, 95% CI: 1.03-1.32, p=0.01). Worse OS was also associated with the higher stage (HR: 3.7, 95% CI: 1.6-8.7, p=0.003), non-dysgerminoma histology (HR: 4.8, 95% CI: 1.1-20.0, p=0.04), and older age (HR: 1.28 per 5-year increase, 95% CI: 1.09-1.50, p=0.002). <b>Conclusions:</b> A female-specific mIGCCCG risk score identifies good-risk pts with better outcomes who may benefit from less intensive chemotherapy than intermediate- and poor-risk pts requiring more intensive chemotherapy and surveillance.Fig. 1