Baseline predictors of early treatment failure in patients with platinum resistant/refractory (PRR) and potentially platinum sensitive (PPS ≥ 3) recurrent ovarian cancer (ROC) receiving ≥ 3 lines of chemotherapy: The GCIG Symptom Benefit Study (SBS).

TitleBaseline predictors of early treatment failure in patients with platinum resistant/refractory (PRR) and potentially platinum sensitive (PPS ≥ 3) recurrent ovarian cancer (ROC) receiving ≥ 3 lines of chemotherapy: The GCIG Symptom Benefit Study (SBS).
Publication TypeJournal
Year of Publication2015
AuthorsRoncolato, FT, O'Connell, R, Joly, F, Lanceley, A, Hilpert, F, Okamoto, A, Aotani, E, Pignata, S, Donnellan, PP, Oza, AM, Åvall-Lundqvist, E, Berek, JS, Sjoquist, KMarie, Gillies, K, Butow, P, Stockler, MR, King, MTrudy, Friedlander, MLeonard
Other NumbersBoard #122; J Clin Oncol 33, 2015 (suppl; abstr 5564)
Keywordsovarian cancer
Abstract

Background: Women with PRR/PPS ≥ 3 ROC are a heterogeneous group with unpredictable response to palliative chemotherapy (PC). GCIG SBS recently completed recruitment of 949 patients treated with PC. Primary aim is to validate an instrument to measure symptom benefit; secondary aims include identifying factors that predict early progression. 25% of patients with PRR-ROC received < 8 weeks of PC. Methods: Physicians recorded baseline characteristics, symptoms (symptomatic ascites, cramping abdominal pain), site/extent of disease and prespecified lab values. Association between baseline characteristics and progression-free survival (PFS) was assessed using time-to-event methods. Median PFS was calculated according to clinically relevant categories and log-rank test applied to assess prognostic value. Cox regression was used to compute hazard ratios and 95% CI to assess the effect of variables on PFS. Results: Sufficient follow up for analysis of PFS was available in 791 patients. Median PFS and overall survival were 4.3 (95% CI: 3.9-4.9) and 12.9 months (95% CI: 11.4-14.0) respectively. In univariate analysis factors with statistically significant associations with PFS included: haemoglobin, PRR-ROC, ascites and abdominal cramps, nodal disease, thrombocytosis, CA125 > 1000, LDH > 600, ECOG status, and elevated c reactive protein. Non-significant factors included: visceral metastases, albumin < 25, lymphocytes < 0.5, tumour volume. Significant variables in multivariable analysis included: ECOG ≥ 2 (HR 1.61 95% CI 1.18-2.19 p = 0.003); nodal disease (HR 1.37 95%CI 1.13-1.67 p = 0.002); ascites (HR 1.54 95%CI 1.24-1.92 p = 0.0001); platinum resistant vs. sensitive (HR 1.39 95%CI 1.12-1.72 p = 0.002), CA125 > 1000 (HR 1.35 95%CI 1.09-1.67 p = 0.005); LDH > 600 (HR 1.88 95%CI 1.36-2.60 p = 0.0001). Conclusions: Several simple clinical variables help predict patients who progress rapidly and will be used to construct prognostic models to aid clinical decisions and trial stratification for clinical trials in PRR/PPS ≥ 3 ROC Clinical trial information: 12607000603415.

URLhttp://meetinglibrary.asco.org/content/151091-156