Abstract
HER2, encoded by the ERBB2 gene, is an important druggable driver of human cancer gaining increasing importance as a therapeutic target in urothelial carcinoma (UC). The genomic underpinnings of HER2 overexpression in ERBB2 nonamplified UC are poorly defined. To address this knowledge gap, we investigated 172 UC tumors from patients treated at the University of California San Francisco, using immunohistochemistry and next-generation sequencing. We found that GATA3 and PPARG copy number gains individually predicted HER2 protein expression independently of ERBB2 amplification. To validate these findings, we interrogated the Memorial Sloan Kettering/The Cancer Genome Atlas (MSK/TCGA) dataset and found that GATA3 and PPARG copy number gains individually predicted ERBB2 mRNA expression independently of ERBB2 amplification. Our findings reveal a potential link between the luminal marker HER2 and the key transcription factors GATA3 and PPARG in UC and highlight the utility of examining GATA3 and PPARG copy number states to identify UC tumors that overexpress HER2 in the absence of ERBB2 amplification. In summary, we found that an increase in copy number of GATA3 and PPARG was independently associated with higher ERBB2 expression in patient samples of UC. This finding provides a potential explanation for HER2 overexpression in UC tumors without ERBB2 amplification and a way to identify these tumors for HER2-targeted therapies.
Keywords: HER2, urothelial cancer, GATA3, PPARG, ERBB2 amplification
This study found that an increase in copy number of GATA3 and PPARG was independently associated with higher ERBB2 expression in patient samples of urothelial carcinoma, providing a potential explanation for HER2 overexpression in tumors without ERBB2 amplification and a way to identify these tumors for HER2-targeted therapies.
Urothelial carcinoma (UC) is an important cause of cancer morbidity and mortality worldwide. HER2 (encoded by ERBB2) is an oncogenic driver frequently altered in multiple cancer types, including breast cancer and malignancies of the gastrointestinal tract,1 where HER2-targeted therapies have improved patient outcomes and have received regulatory approvals. ERBB2 is amplified in approximately 7% of UC tumors with correspondingly high HER2 expression on immunohistochemistry (IHC).2 A recent study reported HER2 positivity (IHC 3+ or 2+ with ERBB2 amplification by fluorescent in situ hybridization) in 15% of UC tumors, while approximately 50% of tumors showed some degree of HER2 expression (IHC ≥ 1+).2 Ongoing clinical trials are investigating the efficacy of HER2-targeted antibody-drug conjugates (ADCs) in UC with promising results, using HER2 protein expression as the relevant biomarker.3,4 This has led to the recent FDA approval of trastuzumab deruxtecan for patients with unresectable or metastatic HER2-positive (IHC 3+) solid tumors, including UC. However, aside from ERBB2 amplification,2 the genetic basis of HER2 overexpression in UC is not well understood.
To assess HER2 expression in UC, we retrospectively identified 172 tumor samples of advanced UC (upper tract and muscle-invasive bladder cancer [MIBC]) from 162 patients treated at the University of California San Francisco (UCSF) and sequenced using UCSF500 (an institutional sequencing panel) and evaluated their HER2 expression using IHC (Supplementary Methods). Among the 162 patients, 87 (53.7%) developed metastatic disease during their clinical courses, 32 (19.8%) had upper tract primary tumors, 116 (71.6%) had definitive surgery, and 9 (5.6%) received definitive chemoradiation. The 87 patients with metastatic disease received a median of 2 lines of systemic therapy in the metastatic setting. Using the gastric algorithm5 for interpreting HER2 IHC (Ventana, Tuscon, AZ; clone 4B5, ready to use), we found HER2 IHC scores of 0, 1+, 2+, and 3+ in 30.2%, 20.3%, 32%, and 17.4% of samples, respectively (Figure 1A). A higher proportion of IHC 3+ was observed in metastatic versus primary tumors (29.8% vs 11.3%; P = .005; Supplementary Figure S1).
Figure 1.
GATA3 and PPARG copy number gains are independently associated with increased ERBB2 expression in urothelial carcinoma. (A) Distribution of HER2 IHC scores of 172 urothelial carcinoma (UC) samples in the UCSF cohort. (B) ERBB2 genetic alterations detected on UCSF500, the institutional cancer gene panel. CNV, copy number variant; SNV, single nucleotide variant. (C) ERBB2 amplifications are enriched in HER2 IHC 3+ UC tumors in the UCSF cohort. (D, E) GATA3 and PPARG amplifications are each enriched in HER2 IHC3+ UC tumors in the UCSF cohort. P values were calculated using Fisher’s exact test comparing the proportions of events (ie, GATA3 and PPARG amplifications, respectively) in HER2 IHC 3+ versus 0/1+/2+ samples. (F, G) GATA3 and PPARG gains or amplifications (as defined by the study) are each enriched in muscle-invasive bladder cancer (MIBC) tumors with high ERBB2 mRNA expression (>4th quartile vs ≤4th quartile) in the MSK/TCGA cohort.
To define the genomic profiles of HER2-overexpressing UC, we first examined ERBB2 alterations (Figure 1B) and found that ERBB2 amplification was present in 12/30 (40.0%) of HER2 IHC 3+ versus 4/142 (2.8%) of HER2 IHC 0/1+/2+ tumors (Figure 1C; p = 1.1 × 10−7). Next, to investigate the gene dosage effect for other cancer-related genes (genes included on the UCSF500 panel other than ERBB2) on HER2 expression, we evaluated a logistic regression model to predict HER2 IHC 3+ (vs IHC 0/1+/2+) status while adjusting ERBB2 amplification status as a covariate (Supplementary Methods). We were able to test 500 and 524 cancer-related genes for amplification and deletion, respectively (Supplementary Tables S1 and S2). We identified 46 gene amplifications and 32 gene deletions associated with HER2 IHC 3+ status with a nominal P < .05 and searched for their overlaps with the top 200 most frequently amplified and deleted genes, respectively, in the Memorial Sloan Kettering/The Cancer Genome Atlas (MSK/TCGA) MIBC dataset (Supplementary Table S3).6 This analysis identified 2 genes, GATA3 and PPARG, both encoding key transcription factors (TFs) in urothelial development and carcinogenesis7 and frequently amplified in UC,8 that were enriched in HER2 IHC 3+ tumors (Supplementary Table S1; Figure 1D, 1E). Among the 500 genes tested for amplification, GATA3 was the second (100th percentile) and PPARG the 45th (91st percentile) most significant gene independently associated with HER2 IHC 3+ status in the UCSF dataset (Supplementary Table S1). The remaining 44 genes with amplifications and all 32 genes with deletions were not among the top 200 most frequently amplified and deleted genes in the MSK/TCGA dataset, respectively (Supplementary Figure S2A, S2B).
To further examine the GATA3 and PPARG gene dosage effects on ERBB2 expression, we interrogated the MSK/TCGA MIBC dataset with concurrent DNA and RNA sequencing data accessible through cBioPortal6,8 (Supplementary Methods). Because HER2 IHC data were unavailable, we analyzed ERBB2 mRNA expression as related to GATA3 and PPARG copy number states. We observed higher ERBB2 mRNA expression in tumors with higher GATA3 and PPARG copy numbers, in all samples (n = 296; P = .0067 and 1.4 × 10−10, respectively; Supplementary Figure S3A, S3B) and in ERBB2 nonamplified samples (n = 280; P = .023 and 3.1 × 10−9, respectively; Supplementary Figure S3C, S3D). Higher GATA3 and PPARG copy numbers were also each associated with higher mRNA expression of the corresponding genes (P = .0021 and <2.2 × 10−16, respectively; Supplementary Figure S3E, S3F). GATA3 and PPARG copy number states were each independently associated with ERBB2 mRNA expression (P = .009 and 3.3 × 10−6, respectively) in a multivariable linear regression model with ERBB2 copy number state as a covariate (Supplementary Methods). ERBB2 mRNA expression was strongly correlated with GATA3 and PPARG mRNA expression, respectively (ρ = 0.55, P = 1.4 × 10−24 and ρ = 0.55, P = 5.2 × 10−25, respectively; Supplementary Figure S4A, S4B). Further, we observed enrichment of GATA3 and PPARG copy number gains in MIBC tumors with the top quartile of ERBB2 mRNA expression, in all samples (n = 296, P = .0006 and 2.1 × 10−10, respectively; Figure 1F, 1G) and ERBB2 nonamplified samples (n = 280, P = .0028 and 4.2 × 10−9, respectively; Supplementary Figure S5A, S5B).
GATA3 and PPARG encode key TFs driving luminal UC7 and are frequently amplified in MIBC (Supplementary Figure S6A, S6B). HER2 is a marker of luminal UC,7,9 particularly the luminal unstable (LumU) subtype.7 In ReMap,10 DNA-binding signals of GATA3 and PPARG (Supplementary Figure S7A, S7B) were observed at the ERBB2 promoter in multiple cell lines, suggesting potential involvement of GATA3 and PPARG in transcriptional regulation of ERBB2, although further studies in UC are needed to test this hypothesis.
To assess the clinical utility of examining gene dosage of GATA3 and PPARG for identifying HER2-overexpressing UC tumors, we first analyzed the UCSF cohort. While 12 of the 30 (40.0%) HER2 IHC 3+ tumors had ERBB2 amplification (Figure 1C), 8 more HER2 IHC 3+ tumors (8/30; 26.7%) without ERBB2 amplification were identified by examining the copy number of GATA3 and PPARG, including 5 with GATA3 amplification alone, one with PPARG amplification alone, and 2 with both GATA3 and PPARG amplifications (Figure 2A; Supplementary Figure S8). We next evaluated the MSK/TCGA cohort, where ERBB2 mRNA levels were used as a surrogate for HER2 expression due to a lack of HER2 IHC data. We found that assessing gene dosage of GATA3 and PPARG in addition to ERBB2 consistently identified more tumors overexpressing ERBB2 mRNA (Figure 2B). For example, among MIBC tumors in the top quartile of ERBB2 mRNA expression, 20.3% displayed ERBB2 amplification, but analyzing for GATA3 or PPARG amplification identified an additional 24.3% of tumors overexpressing ERBB2 mRNA. Thus 44.6% of all MIBC tumors in the top quartile of ERBB2 mRNA expression harbored amplification involving at least one of ERBB2, GATA3, and PPARG. This relationship was consistent across all levels of ERBB2 mRNA expression (Figure 2B).
Figure 2.
Evaluating GATA3 and PPARG amplifications identifies more tumors with ERBB2 overexpression. (A) In the UCSF cohort, GATA3 and PPARG amplifications help identify 8 additional HER2 IHC 3+ tumors without ERBB2 amplification. (B) In the MSK/TCGA cohort, consistently more tumors with ERBB2 mRNA overexpression are identified by evaluating the amplification status of GATA3 and PPARG in addition to ERBB2 itself.
In summary, we reported HER2 overexpression in UC tumors with GATA3 and PPARG copy number gains and demonstrated the clinical utility of examining these 2 genes for identifying additional UC tumors overexpressing HER2, particularly when ERBB2 amplification is not detected. In addition to providing a potential explanation for HER2 overexpression in non-ERBB2 amplified UC tumors, our findings emphasize the need to look beyond ERBB2 amplification when using a tumor DNA-based assay to help assess HER2 expression status in UC, an important biomarker for novel HER2-targeted ADCs.
Supplementary Material
Acknowledgments
This study was funded by an intramural award at the Helen Diller Family Comprehensive Cancer Center sponsored by The Cancer League. The title of the funded project was “Clinicogenomic profiles and treatment outcomes of HER2-expressing and HER2-altered urothelial cancer” and the principal investigators were Dr. Vadim Koshkin and Dr. Xiaolin Zhu.
Contributor Information
Xiaolin Zhu, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Medicine, University of California San Francisco, San Francisco, CA, United States.
Emily Chan, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Pathology, Stanford Medical Center, Stanford, CA, United States.
Michelle L Turski, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States.
Carlos Espinosa Mendez, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States.
Sarah C Hsu, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Medicine, University of California San Francisco, San Francisco, CA, United States.
Vipul Kumar, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Medicine, University of California San Francisco, San Francisco, CA, United States.
Chase Shipp, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Medicine, University of California San Francisco, San Francisco, CA, United States.
Tanya Jindal, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Medicine, University of California San Francisco, San Francisco, CA, United States.
Kevin Chang, School of Medicine, University of California San Francisco, San Francisco, CA, United States.
Courtney Onodera, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States.
W Patrick Devine, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Pathology, University of California San Francisco, San Francisco, CA, United States.
James P Grenert, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Pathology, University of California San Francisco, San Francisco, CA, United States.
Bradley A Stohr, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Pathology, University of California San Francisco, San Francisco, CA, United States.
Chien-Kuang Cornelia Ding, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Pathology, University of California San Francisco, San Francisco, CA, United States.
Matthew D Stachler, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Pathology, University of California San Francisco, San Francisco, CA, United States.
David A Quigley, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Urology, University of California San Francisco, San Francisco, CA, United States.
Felix Y Feng, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Radiation Oncology, University of California San Francisco, San Francisco, CA, United States.
Carissa E Chu, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Urology, University of California San Francisco, San Francisco, CA, United States.
Sima P Porten, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Urology, University of California San Francisco, San Francisco, CA, United States.
Jonathan Chou, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Medicine, University of California San Francisco, San Francisco, CA, United States.
Terence W Friedlander, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Medicine, University of California San Francisco, San Francisco, CA, United States.
Vadim S Koshkin, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, United States; Department of Medicine, University of California San Francisco, San Francisco, CA, United States.
Conflicts of interest
V.S.K. has served in a consulting or advisory role for AstraZeneca, Janssen, Pfizer, EMD Serono, Seagen, Astellas, Guidepoint, GLG and ExpertConnect; has received research funding for the institution from Endocyte/Novartis, Nektar, Gilead, Janssen, Taiho, Merck and Seagen, individual research funding from Eli Lilly and is supported by the Prostate Cancer Foundation. No other authors report competing interests related to this work.
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
References
- 1. Oh D-Y, Bang Y-J.. HER2-targeted therapies - a role beyond breast cancer. Nat Rev Clin Oncol. 2020;17(1):33-48. 10.1038/s41571-019-0268-3 [DOI] [PubMed] [Google Scholar]
- 2. Scherrer E, Kang A, Bloudek LM, Koshkin VS.. HER2 expression in urothelial carcinoma, a systematic literature review. Front Oncol. 2022;12:1011885. 10.3389/fonc.2022.1011885 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Zhou L, Xu H, Li S, et al. Study RC48-C014: preliminary results of RC48-ADC combined with toripalimab in patients with locally advanced or metastatic urothelial carcinoma. J Clin Oncol. 2022;40(6_suppl):515-515. 10.1200/jco.2022.40.6_suppl.515 [DOI] [Google Scholar]
- 4. Meric-Bernstam F, Makker V, Oaknin A, et al. Efficacy and safety of trastuzumab deruxtecan in patients with HER2-expressing solid tumors: primary results from the DESTINY-PanTumor02 phase II trial. J Clin Oncol. 2024;42(1):47-58. 10.1200/JCO.23.02005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Hofmann M, Stoss O, Shi D, et al. Assessment of a HER2 scoring system for gastric cancer: results from a validation study. Histopathology. 2008;52(7):797-805. 10.1111/j.1365-2559.2008.03028.x [DOI] [PubMed] [Google Scholar]
- 6. Pietzak EJ, Zabor EC, Bagrodia A, et al. Genomic differences between “primary” and “secondary” muscle-invasive bladder cancer as a basis for disparate outcomes to cisplatin-based neoadjuvant chemotherapy. Eur Urol. 2019;75(2):231-239. 10.1016/j.eururo.2018.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kamoun A, de Reyniès A, Allory Y, et al. ; Bladder Cancer Molecular Taxonomy Group. A consensus molecular classification of muscle-invasive bladder cancer. Eur Urol. 2020;77(4):420-433. 10.1016/j.eururo.2019.09.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Gao J, Aksoy BA, Dogrusoz U, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6(269):l1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Choi W, Porten S, Kim S, et al. Identification of distinct basal and luminal subtypes of muscle-invasive bladder cancer with different sensitivities to frontline chemotherapy. Cancer Cell. 2014;25(2):152-165. 10.1016/j.ccr.2014.01.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hammal F, de Langen P, Bergon A, Lopez F, Ballester B.. ReMap 2022: a database of human, mouse, drosophila and Arabidopsis regulatory regions from an integrative analysis of DNA-binding sequencing experiments. Nucleic Acids Res. 2022;50(D1):D316-D325. 10.1093/nar/gkab996 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.