Outcomes with tandem or triplet autologous stem cell transplant for refractory germ cell tumors.

632 Background: While Germ Cell Tumor (GCT) remain highly curable, up to 40% of patients (pts) with intermediate or poor risk advanced GCT will relapse. In the relapsed setting, high dose chemotherapy followed by autologous hematopoietic cell transplant (AHCT) may be pursued with curative intent. To predict the risk of post-AHCT relapse, the International Prognostic Factors Study Group created 5 risk categories incorporating 7 variables. The IPFSG score was created prior to the popularization of tandem/triplet transplants and validated in patients coming to AHCT in first relapse. Methods: We performed a retrospective analysis of all pts with GCT treated with AHCT at Stanford from 2017-2024. Pts were included if they were planned for doublet "tandem" (N Engl J Med 2007;357:340) or triplet (J Clin Oncol 2010;28:1706) AHCT with at least two months of follow up. Baseline characteristics, treatment history and outcomes were analyzed. For pts who came to transplant after >1 relapse, their IPFSG score was calculated based on most recent therapy prior to AHCT. Median relapse free survival (mRFS), overall survival (mOS), and 1-year outcomes were calculated from the final transplant and stratified by IPFSG score. A cox proportional analysis was performed to evaluate the multivariate for age, site of primary tumor, presence of brain metastases, and number of relapses prior to AHCT. Results: 57 pts were included in the analysis. Median age was 29 years (range 18-52). All pts were mobilized with G-CSF and underwent conditioning with carboplatin and etoposide. Three (5%) pts were planned for tandem and 54 (95%) for triplet AHCT. Most (97%) were non-seminomas or mixed germ cell tumors. 12 (21%) pts were transplanted after >1 relapse. The disease status at time of AHCT was partial remission tumor marker (PRM) negative in 9 (16 %), PRM positive in 43 (75%) and progressive disease in 5 (9%). 13 (23%) pts did not complete all scheduled transplants, all due to progression of disease. The median follow-up for the entire cohort was 330 days. The mRFS was 115 days, and mOS was 344 days. For the 49 pts at least one year from transplant, one-year RFS and OS are shown in the table. In a multivariate analysis, there was no association between death and age, site of primary tumor, presence of brain metastasis, disease status at transplant, or number of recurrences. Conclusions: Long term outcomes for patients with refractory GCT remain poor, though a subset of patients may still be cured with AHCT. The IPFSG accurately prognosticates outcomes in our real-world analysis. Outside of their contribution to IPFSG, variables such as disease status at transplant, site of metastasis, and number of recurrences did not have statistical significance. 1 year RFS and OS by IPFSG score. IPFSG Score Very low Low Intermediate High Very High Overall 1 yr RFS 100% (3/3) 66% (2/3) 78% (7/9) 20% (5/25) 11% (1/9) 37% (18/49) 1 yr OS 100 (3/3) 100% (3/3) 89% (8/9) 44% (11/25) 33% (3/9) 57% (28/49)