Page 116 - Read Online

Page 22 of 26 Li et al. Cancer Drug Resist. 2025;8:31

PCa is a hormone-dependent malignant tumor, and hormone receptor pathway-targeted therapies are the
primary approaches to treatment. Second-generation androgen receptor inhibitors, such as apalutamide,
have extensive clinical use. Although these drugs have improved the survival of patients with PCa, issues
related to drug resistance remain a substantial challenge. Our findings suggest that different cancer cell
subgroups exhibit varying levels of resistance to different anticancer drugs. Only subgroup 3 cancer cells
were sensitive to apalutamide treatment, indicating the need to tailor treatment strategies based on the
distribution of cancer cell subgroups. Future work will prioritize expanding the drug panel to include
additional clinically critical therapeutics for mechanistic interrogation. Key resistance mechanisms identified
herein will undergo rigorous validation in physiologically relevant preclinical models and/or clinical cohorts.
Overall, we believe that interfering with cancer cell energy metabolism pathways using drugs that target
specific metabolic enzymes might represent a potential therapeutic strategy. ATP-binding cassette (ABC)
transporters, including ABCB1 and ABCG2, are recognized contributors to chemotherapy resistance .
[51]
However, our multi-omics analysis and genome-wide modeling did not identify these transporters as
primary drivers of the immune-related risk subgroups. This indicates that within the immune-associated
high-risk subtype identified in this study, resistance mechanisms are more likely influenced by dynamic
changes in the immune microenvironment rather than conventional drug efflux systems. Our findings
underscore the heterogeneity of drug resistance pathways in PCa and emphasize the importance of
differentiating between immune-mediated and classical efflux-associated resistance mechanisms, which have
significant implications for personalized treatment strategies.

Although resistance analyses have been conducted on different cancer cell subtypes, the efficacy of anticancer
drugs in various clinical cancer cell subtypes remains a subject for further investigation. It should be noted
that metabolic pathway identification in this study primarily relied on transcriptomic data, without
orthogonal validation methods like metabolomics, which may affect result accuracy. In the future, we plan to
further investigate the relationship between the immune microenvironment and resistance mechanisms, and
integrate metabolomics and other experimental approaches in future work to further validate and refine
metabolic pathway alterations in PCa subtypes. We will integrate additional transcriptomic data and
scRNA-seq data to analyze drug-resistant characteristics across distinct immune subtypes. This research will
enhance our understanding of the PCa immune microenvironment and its impact on therapeutic responses
within a broader context, thereby providing novel insights for developing more effective treatment strategies.
Furthermore, we will evaluate the sensitivity of various cellular clusters to specific anticancer agents by
assessing their proliferative capacity. Validation experiments will investigate the differential expression of
resistance-associated genes at both the mRNA and protein levels across these clusters. Cells will be exposed
to gradient concentrations of anticancer drugs to compare IC50 values among subpopulations, thus
analyzing the heterogeneity in drug resistance. Importantly, metabolic intervention may serve as a critical
breakthrough based on our identified gene signatures. For instance, inhibiting the glycolytic pathway in
tumor cells could attenuate their survival capacity. Notably, we observed a significant enrichment of muscle
cell pathways, and we posit that this phenomenon may reflect distributional differences in stromal
components (e.g., myofibroblasts) or processes such as epithelial-mesenchymal transition across risk
subtypes within tumor tissues, although current evidence does not provide direct mechanistic support. Given
the complexity of multi-omics data and the heterogeneity of the TME, the biological significance of this
observation warrants further investigation through subsequent experiments and higher-resolution studies.
We therefore present this finding as a novel observation that merits attention, and recommend future
research to elucidate its underlying mechanisms and prognostic implications in PCa. We integrated drug
resistance data from the GDSC database with PCa cell line expression profiles. These findings should be
regarded as hypothesis-generating, offering preliminary insights that require validation through further
experimentation and clinical investigation. While different cell subpopulations exhibited variable tolerability
and sensitivity profiles to the four anticancer agents, suggesting potential differential optimal drug selection

109