Prognostic Differences Between Surveillance and Active Treatment After Initial Orchiectomy in Patients With Stage I Mixed Germ Cell Tumors of the Testis: A Propensity Score Matching Analysis

Abstract

IntroductionThe prognosis and optimal treatment approach for stage I mixed germ cell cancers of the testis are not well-established. This study aimed to assess contemporary treatment rates and their correlation with the cancer-specific mortality (CSM) and other-cause mortality (OCM) in patients with stage I testicular mixed germ cell tumors (TMGCT) who underwent orchiectomy, comparing surveillance with active treatment, including chemotherapy (CHT) and retroperitoneal lymph node dissection (RPLND). MethodsRetrospective analysis of clinical data from stage I TMGCT patients who underwent orchiectomy was conducted using the Surveillance, Epidemiology, and End Results database from 2004 to 2019. The annual percentage change (APC) in the use of surveillance, postoperative CHT, and RPLND was examined. Propensity score matching (PSM) and cumulative incidence, analyses were employed to compare differences in CSM and OCM between surveillance and active treatment, as well as between CHT and RPLND. Multivariate competing-risks regression models were utilized to investigate independent factors affecting CSM and OCM among stage I TMGCT patients. ResultsThe study included 5743 individuals with stage I TMGCT that underwent surveillance (61.6%), CHT(27.2%), or RPLND (11.2%). Among them, 82 deaths were attributed to TMGCT, and 82 deaths resulted from other causes. Surveillance rates increased over time (APC: 0.635%, P = 0.008), as did CHT rates (APC: 0.863%, P < 0.001), while RPLND rates declined (APC: −0.96%, P < 0.001). After PSM, multivariate competing-risks regression analysis showed that, active treatment, compared to surveillance, was not an independent factor for CSM and OCM. In contrast, when compared to CHT, RPLND was an independent factor associated with lower CSM (hazard ratio = 0.247, 95% confidence interval: 0.08-0.761; P = 0.015), but not OCM (hazard ratio = 0.946, 95% confidence interval: 0.377-2.37; P = 0.91). ConclusionsSurveillance and CHT rates have increased over time for patients with stage I TMGCT following initial orchiectomy, while RPLND utilization has decreased. There was no significant difference in CSM between surveillance and active treatment groups, but RPLND demonstrated significantly lower CSM than CHT in active treatment. Our findings suggest that the usage of RPLND in patients with stage I TMGCT should be reconsidered.