859P - Clinical outcomes of poor prognosis germ cell tumors (GCT): An analysis of a series from a single referral center

Date 27 September 2014
Event ESMO 2014
Session Poster Display session
Topics Germ Cell Tumours
Presenter Elena Farè
Citation Annals of Oncology (2014) 25 (suppl_4): iv280-iv304. 10.1093/annonc/mdu337
Authors E. Farè1, S. Lo Vullo2, P. Giannatempo1, N. Nicolai3, D. Raggi1, L. Piva4, D. Biasoni4, M. Catanzaro4, T. Torelli4, M. Marongiu1, S. Stagni4, M. Maffezzini4, L. Mariani5, A.M. Gianni6, R. Salvioni3, A. Necchi1
  • 1Medical Oncology/urology Unit, Fondazione IRCCS - Istituto Nazionale dei Tumori, 20133 - Milano/IT
  • 2Clinical Epidemiology And Trials Organization Unit, Fondazione IRCCS - Istituto Nazionale dei Tumori, Milano/IT
  • 3Surgery Urology Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, 20133 - Milan/IT
  • 4Surgery - Urology Unit, Fondazione IRCCS - Istituto Nazionale dei Tumori, 20133 - Milano/IT
  • 5Clinical Epidemiology And Trials Organization Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, 20133 - Milano/IT
  • 6Medical Oncology, University of Milan - Fondazione IRCCS Istituto Nazionale dei Tumori, 20133 - Milan/IT

Abstract

Aim

Outcomes of first-line treatment (Tx) for patients (pts) with metastatic poor prognosis GCT are still suboptimal in literature. However, the low incidence of disease and the increasing effectiveness of salvage therapy are issues against further clinical research in the field. We conducted a retrospective study to evaluate the impact of Tx period on prognosis of pts referred to our tertiary cancer center.

Methods

A retrospective analysis was conducted on pts receiving at least first-line chemotherapy (CT) at our center. Distribution of frequencies of clinical characteristics were evaluated in the periods <1997, 1997-2001, 2001-06, and 2007-13. The Kaplan-Meier method was used to estimate progression-free (PFS) and overall survival (OS). PFS and OS were calculated since the start of first-line CT. Tx period was included in univariable (UVA) and multivariable (MVA) Cox regression models for PFS and OS together with the following variables: tumor primary, liver-bone-brain (LBB) metastases, first-line PEB vs high-dose CT (HDCT), AFP>1000 IU/ml, HCG > 1000 IU/l. All tests and confidence intervals were two-sided and set at p = 0.05 level of significance.

Results

Between the years 1982 and 2013, 168 pts have been identified. Median age was 27 yrs (IQR: 22-34). No clinical factor had significantly different distribution over time, only LBB metastases trended to higher frequency from 1997 onwards (27.5% <1997 to 55.6% in 2007-2013, Chi-squared test p = 0.054). Median follow up was 102 months (IQR: 63-166). Global 5-year PFS was 48.5% (95%CI, 41.5-56.8) and OS was 63.2% (95%CI, 56.0-71.2). On UVA for PFS, Tx period 2007-2013 vs <1997 trended to lower survival (HR: 1.76, 95%CI: 0.97-3.21, p = 0.063). On MVA, Tx period was not significantly associated with neither PFS (HR: 1.72, 95%CI: 0.87-3.41, p = 0.229) nor OS (HR: 0.66, 95%CI, 0.27-1.59, p = 0.216).

Conclusions

In this single-center series of poor prognosis GCT we could not demonstrate an effect on Tx period on survival. The slight differences in PFS on UVA are likely due to modest discrepancies in distribution of clinical variables (LBB). The global picture is that of a stable and very high cure-rate. Based on these results, attempts to improve the outcome exclusively in this field should be discouraged. Results are biased by their retrospective quality.

Disclosure

All authors have declared no conflicts of interest.