Introduction

Tumor heterogeneity has been a major contributor to lethal outcomes, drug resistance, and therapeutic failures, and presents a key challenge to precision medicine goals1,2. In light of the association between tumor heterogeneity and poor prognostic results, a measure of heterogeneity itself might be useful as a prognostic marker3. Although heterogeneity of prostate cancer is conventionally attributed to genetic diversity4,5,6, current evidence reveals that in addition to genetic factors, the tumor heterogeneity could be derived from nongenetic variabilities7.

The stroma serves as a main barrier preventing carcinogenesis in benign tissue; however, the presence of cancer cells initiates crucial changes, converting the environment into one that supports tumor growth. These changes include fibroblast recruitment, immunocytes migration, matrix remodeling, development of tumor-specific vasculature, and aberrant epigenetic landscape, and each of these changes might promote heterogeneity of tumor microenvironment (TME). Local diversity of selective pressures in TME, such as hypoxia, acidity, and growth factors, also actively shape tumor morphology. Conceivably, the distinct environmental landscape in the tumor plays a significant role in tumor heterogeneity.

TME is dynamic with spatial and temporal changes in composition in response to environmental pressures and anticancer therapies, and the continued crosstalk between tumor cells and the surrounding microenvironment is fundamental to tumor initiation, phenotypic changes, cancer progression, and therapeutic resistance. In this review, we summarize the challenges targeting the tumor environments, and emphasize that understanding nongenetic mechanisms might open novel diagnostic and therapeutic approaches with the potential to improve the efficacy of current prostate cancer treatments.

Clonal heterogeneity

Prostate cancer is a multifocal disease (Fig. 1) and each focus might have a different phenotype (intertumor heterogeneity)8. As the individual tumor volume increases over time, multiple foci might merge into a larger mass exhibiting greater tumor heterogeneity (intratumor heterogeneity). It is controversial whether those separate foci reflect a monoclonal origin9,10,11,12 or a polyclonal origin13. The studies supporting the monoclonal origin argue that both genetic and epigenetic events occur in a single ancestral cell that does not possess all the necessary mutations to transform into a cancer cell. As those cells are exposed to additional events and divide, DNA replication errors lead to daughter cells that are genetically different from each other. Hence, although the cells are heterogeneous, all derive from the same ancestor14. For instance, analysis of the whole exome sequencing and transcriptome profiles from Gleason 3 and neighboring Gleason 4 tumor foci revealed that the adjacent tumors emerged early from a common precursor and subsequently undergo independent evolution10. In a separate study, analysis of genomewide single nucleotide polymorphism and copy number variations from metastasized prostate cancers demonstrated that most cancers were of monoclonal origin and have identical copy number changes11.

Fig. 1: Intratumor and intertumor heterogeneity of prostate cancer.
figure 1

Whole-mount cross-section of a radical prostatectomy specimen has two separated tumor foci. One focus is in the right anterior of the prostate (Gleason score 4 + 3 = 7, Group 3) whereas another focus is in the left posterior of the prostate (Gleason score 5 + 4 = 9, Group 5). Scale bars, 4 mm (left) and 100 µm (right). Methods: Radical prostatectomy specimens were serially sectioned into 3 mm slices and completely embedded. The case was reviewed by a single urologic pathologist (R.G.) in 2021. The following features were monitored: Gleason Score and Grade Group according to the International Society of Urological Pathology (ISUP) 2014 guidelines. The percentages of Gleason pattern 3, 4, and 5 were estimated, including presence of tertiary Gleason patterns.

In support of the polyclonal tumor origin, the chromosomal analysis and genomic DNA sequencing studies on multifocal prostate cancers have revealed that different tumor foci have independent clonal expansions5,13,15. In another study, analysis of the whole exome sequencing of 23 distinct tumor foci from 5 prostate cancer cases demonstrated that the multifocal tumors were highly heterogeneous for single nucleotide variants, copy number aberrations, and genomic rearrangements16. Similarly, a large cohort study of 89 tumor foci from 41 different patients revealed that samples from different tumor foci in the same prostatectomy specimen rarely had any shared point mutations and the same DNA copy number changes8.

Moreover, analysis of 17 tumor cells from localized lesions with different Gleason scores from 2 prostatectomy cases revealed that in patient number 1, every cell had the same TP53 mutation, which is consistent with the monoclonal model. In the patient number 2, only one cell subpopulation contained the TP53 mutation, while other cells carried different mutations, supporting a polyclonal model17. Overall, these studies indicated that the origin of prostate cancers may have a monoclonal or polyclonal origin that varies from case to case. Further research may shed more light on the generality or predominance of any of these theories.

Tumor microenvironment heterogeneity

It is well-accepted that tumorigenesis is not only dependent on genetic alterations or epigenetic modifications in cancer cells, but is also regulated by the TME18. The TME is composed of fibroblasts, pericytes, immunocytes, and endotheliocytes; each able to crosstalk with cancer cells in dynamic ways (Fig. 2). Usually, the orchestrated impact of microenvironmental components on cancer cells is characterized by the different region, and the tumorigenesis is modulated by the regional heterogeneity in the hypoxia, acidity, and cytokines in a tumor environment19,20. Moreover, the cancer-associated fibroblasts (CAFs) are among the most abundant constituents in the TME, contributing to the malignant phenotype at all levels21,22,23,24,138. Tumor heterogeneity may be the reason that biomarkers for treatments targeting the tumor stroma, in general, have remained elusive.

Furthermore, the tumor’s complexity results from ongoing interaction between tumor cells and the environment throughout the course of disease progression, which adds to the challenge of spatiotemporal heterogeneity. Since cancer patients are often treated with multiple lines of therapy, serial biopsies from various locations are necessary to predict treatment responsiveness and make therapeutic decisions.

Conclusions

Heterogeneity of the TME plays a key role in prostate cancer progression. The currently available therapies for prostate cancers, including conventional therapies and immunotherapy, could lead to therapy-induced tumor heterogeneity. Further characterization of genetic and nongenetic heterogeneity will aid in the development of more effective, personalized, and targeting-specific therapies for advanced prostate cancer patients.