2019 - Validation of the Geriatric Vulnerability Score (GVS) in older ovarian cancer (oOC) patients: an analysis from the GCIG-ENGOT-GINECO EWOC-1 study

Background

The GINECO previously developed the GVS to identify geriatric vulnerability in oOC pts (Falandry, 2013). The derivation cohort was EWOT-3 database of 111 oOC pts treated with first line carboplatin. The GVS combines albumin (≥ or < 35g/l), lymphocyte (< or ≥ 1x10⁹/L) levels, and activities of daily living (ADL; ≥ or < 6), Instrumental ADL (≥ or < 25), HADS (< or ≥ 14) scores. With a cut-off ≥3, GVS delineated a population with significantly decreased OS (HR(95%CI)=2.94 (1.79–4.84), P <.0001). EWOC-1 international study included an external validation of the GVS as a secondary endpoint.

Methods

Pts ≥70 yrs diagnosed with FIGO stage III/IV epithelial OC and no organ failure were screened for GVS. Those with GVS≥3 were proposed EWOC-1 randomized trial, evaluating 3 treatment regimens in the vulnerable pts. Other pts’ data were collected in the “EWOC-1 registry”. External validation of GVS was performed in the whole population (V1), in the EWOC-1 registry subgroup (V2), and in the subgroup of pts treated with carboplatin paclitaxel regimens (V3). Cross-validation analyses (calibration, discrimination, and performance analysis) were performed according current recommendations (Steyerber, 2014).

Results

From 12/2013 to 11/2018, 447 elderly patients were included, 120 (27%) in EWOC-1 trial and 327 in EWOC-1 registry (V1: n = 447, V2: n = 327, V3: n = 324). Patients’ cancer characteristics were similar in the validation cohorts compared to the derivation one. Median follow up was 22 mo; missing values were limited (<2%). According the maximum likelihood analysis, the risk ratio of death according GVS score in V1 was (ref=0): 1: 1.8 (p=.029); 2: 2.4 (p<.001); 3: 4.1 (p<.001); 4: 5.5 (p<.001); 5: 9.1 (p<.001). Whatever the validation cohort, GVS≥3 significantly segregated two groups with different OS (median, in mo): V1 (13.2 vs 40.8; HR = 2.8 (2.2,3.7), p<.001), V2 (11.9 vs 40.8, HR = 3.5 (2.5,4.9), p<.001); V3 (18.1 vs 43, HR = 2.6 (1.9,3.7), p<.001).

Conclusions

GVS provides a reproducible and robust prediction model of vulnerability in oOC pts, independent of geographic and historic effect (V1), as well as treatment patterns (V3), validated on an international population.

Clinical trial identification

PROTOCOL EWOC-1 - ENGOT OV-23 - 2013-000266-11.

Editorial acknowledgement

Legal entity responsible for the study

Centre Hospitalier Lyon Sud.

Funding

Hospital Clinical Research Program (PHRC).

Disclosure

D. Lorusso: Honoraria (self), Honoraria (institution), Advisory / Consultancy: MERCK; Advisory / Consultancy: AstraZeneca; Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (self): Clovis; Advisory / Consultancy: Immunogen; Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses, Shareholder / Stockholder / Stock options, Licensing / Royalties, Full / Part-time employment: PharmaMar; Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses, Shareholder / Stockholder / Stock options, Licensing / Royalties, Full / Part-time employment: Roche; Advisory / Consultancy, Travel / Accommodation / Expenses, Shareholder / Stockholder / Stock options, Licensing / Royalties, Full / Part-time employment: Tesaro; Officer / Board of Directors, Spouse / Financial dependant, Non-remunerated activity/ies: GCIG. A. Floquet: Advisory / Consultancy: Clovis; Advisory / Consultancy, Travel / Accommodation / Expenses: AstraZeneca; Advisory / Consultancy, Travel / Accommodation / Expenses: Tesaro; Travel / Accommodation / Expenses: Roche. O. Tredan: Honoraria (self): Roche; Honoraria (self): AstraZeneca; Honoraria (self): Pfizer; Honoraria (self): Novartis; Honoraria (self): Lilly; Honoraria (self): BMS; Honoraria (self): MSD. E. Pujade-Lauraine: Honoraria (self), Honoraria (institution), Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (self), Research grant / Funding (institution), Travel / Accommodation / Expenses, Shareholder / Stockholder /: AstraZeneca; Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (self), Research grant / Funding (institution), Travel / Accommodation / Expenses, Shareholder / Stockholder / Stock options, Licensing / Royalties, Full: Roche; Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (self): Clovis; Honoraria (self), Honoraria (institution), Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (institution), Travel / Accommodation / Expenses, Shareholder / Stockholder / Stock options, Full / Part-time e: Tesaro; Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (self): Genmab; Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (self): Incyte; Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (self): MSD; Advisory / Consultancy, Speaker Bureau / Expert testimony, Leadership role, Research grant / Funding (self): Pfizer. C. Falandry: Honoraria (self): Leo Pharma; Honoraria (self), Honoraria (institution): Pfizer; Honoraria (self): MSD oncology; Honoraria (self): Teva; Honoraria (self): AstraZeneca; Honoraria (self): Baxter; Honoraria (self): Eisai; Honoraria (self): Janssen; Honoraria (self): Novartis; Honoraria (institution): Chugai Pharma; Honoraria (institution): Pierre Fabre; Honoraria (institution): Astellas Pharma. All other authors have declared no conflicts of interest.