Primary mediastinal germ cell tumors (PMGCTs) are a rare type of cancer affecting young adults. They have different molecular and clinical features compared to testicular germ cell tumors. Non-seminoma PMGCTs have the shortest 5-year overall survival and the poorest prognosis among all of the germ cell tumor presentations, while seminomas share the same survival and prognosis as their testicular counterparts. There is an unmet need for better treatment options for patients with non-seminoma PMGCTs in both first-line and salvage therapy, as the available options are associated with underwhelming outcomes. Identifying biological and genetic factors to predict treatment responses would be helpful in improving the survival of these patients.

Purpose: HER2 targeting is increasingly relevant in metastatic urothelial cancer (mUC) because of emerging antibody–drug conjugates, in which assessment of tumor HER2/ERBB2 status could aid interpretation of ongoing trials. We evaluated whether precise ERBB2 genotype can be determined from mUC ctDNA and the relationship with tissue status. Experimental Design: ERBB2 genotype was determined through targeted sequencing of longitudinal samples from 226 patients with mUC. Heterogeneity was evaluated across ctDNA and metachronous tumor tissue ERBB2 genotype and compared with HER2 IHC. Results: Activating ERBB2 mutation and/or amplification was identified in 16% and 29% of patients by ctDNA and tissue sequencing, respectively. Alternatively, 55% of tissue samples were HER2 IHC positive. Agreement between HER2 IHC and ERBB2 genotype (by either ctDNA or tissue) was 64%, in contrast to 87% between patient-matched ctDNA and tissue genotypes. Across serial samples, genotype and IHC revealed marked heterogeneity in ERBB2 status. Factors linked with heterogeneity included ERBB2 amplifications on extrachromosomal DNA, detected through whole-genome sequencing of ctDNA and FISH, and subclonal ERBB2 mutations, which were evident in half of ctDNA and one-third of tissue samples. Conclusions: We report a combined analysis of HER2 IHC and ERBB2 genotype in mUC ctDNA and archival tissue, revealing high spatiotemporal heterogeneity. Our data suggest that sole reliance on DNA- or protein-based HER2 assessment is insufficient to capture nuanced genomic indicators of HER2 pathway reliance and support a role for ctDNA alongside existing methods for characterizing HER2/ERBB2 status during biomarker development in mUC.

PURPOSE: Chemotherapy is the primary treatment for metastatic nonseminomatous germ cell tumors (mNSGCTs), but patients often encounter postchemotherapy residual disease. Accurate noninvasive methods are needed to predict the histology of these masses, guiding treatment and reserving surgery for those with teratoma. This study aims to enhance predictive accuracy by integrating computed tomography (CT) radiomics features with miRNAs (miR371-375) to distinguish between teratoma and nonteratoma histology in postchemotherapy residual masses. METHODS: We retrospectively identified 111 lesions, divided into training and test sets (n = 78 v 33) with equal class distribution. 3D Slicer was used to segment lesions with a short axis of >10 mm from the postchemo-presurgical CT images, and radiomics features were extracted. Presurgery plasma miR371-375 levels were measured by real-time polymerase chain reaction. Four machine learning models evaluated the predictive value of radiomics alone (R-only) and combined with miR371-375 levels, and the best performer was selected. Clinical factors associated with teratoma from univariate analysis were included in multivariate analysis with the best radiomics signature to assess their impact on predicting teratoma histology. RESULTS: The CatBoost (CB) model R + 371 + 375 exhibited the best and most robust overall accuracy for predicting residual teratoma, with the highest AUC values (0.96, 95% CI, 0.88 to 1.0 for training, 0.83, 95% CI, 0.68 to 0.98 for testing) and a well-balanced sensitivity and specificity. Univariate analysis identified presurgery alpha-fetoprotein ( P = .01), beta-human chorionic gonadotropin ( P = .01), initial teratoma pathology ( P = .01), and lymph node metastases ( P = .02) as significant predictors for teratoma. Multivariate analysis included these features and the radiomics signature, which was the strongest independent predictor ( P < .0001). CONCLUSION: Combining miR371-375 with CT radiomics features improves the accuracy of predicting teratoma histology of postchemotherapy residual disease in mNSGCTs and, therefore, has the potential to guide treatment decision making.

Supplementary Table 1. Cohort composition and mUC cfDNA samples.

Supplementary Table 9. ERBB2 structural variants detected by targeted cfDNA sequencing.