The analysis examined data from 4,890 patients treated through the Veterans Health Administration between 2018 and 2023. All had metastatic hormone-sensitive prostate cancer, meaning their disease had spread beyond the prostate but was still responsive to treatments that suppress testosterone and other androgen hormones. Most patients received testosterone deprivation therapy, also known as androgen-deprivation therapy, either alone or in combination with additional medicines designed to block androgen production or prevent testosterone from activating prostate cancer cells.

PSA is a protein produced primarily by prostate cells. Because prostate cancer cells often release PSA into the bloodstream, doctors routinely measure its concentration to monitor how the disease responds to treatment. A declining PSA level generally indicates that the cancer is becoming less active, while a rising level can signal resistance or progression. However, clinicians have used several different ways to interpret a response, including the percentage reduction from a patient’s starting PSA level and whether the concentration reaches a particular absolute value.

In the new study, the researchers focused on whether reaching a PSA level below 0.2 ng/mL was more closely associated with survival than achieving a decline of at least 90 percent. The 0.2 ng/mL threshold has emerged from several phase 3 clinical trials as a potentially meaningful measure of treatment response, but evidence from everyday clinical practice has been more limited. The investigators therefore used health records from a large national medical system to assess how these response measures performed in a community-based, real-world population.

During a median follow-up period of approximately two years, 896 patients died. Those who reached a PSA level below 0.2 ng/mL within nine months of starting treatment were 54 percent less likely to die than patients who did not reach that level. The association remained whether or not the patients also experienced a PSA reduction of at least 90 percent, indicating that the absolute PSA threshold carried prognostic information beyond the size of the decline alone.

The distinction is important because a dramatic percentage reduction does not always mean that the remaining cancer activity is minimal. For example, a patient whose PSA falls from 20 ng/mL to 2 ng/mL has experienced a 90 percent decline, but the final level remains well above 0.2 ng/mL. By contrast, a patient whose PSA drops from 1 ng/mL to 0.15 ng/mL has achieved a smaller percentage reduction but has reached the lower absolute threshold associated with more favorable outcomes in this analysis.

Patients who experienced at least a 90 percent decline but did not reach a PSA level below 0.2 ng/mL showed no significant survival improvement compared with patients who failed to achieve the 90 percent reduction. This finding suggests that the depth of suppression may be more clinically informative than the relative change from baseline. It also supports the idea that PSA response should be assessed using both the starting value and the level that remains after treatment, rather than relying on percentage decline alone.

The study further found that patients treated with testosterone deprivation therapy plus another androgen-blocking drug were more likely to reach a PSA level below 0.2 ng/mL than those receiving hormone suppression without treatment intensification. Modern combination approaches can inhibit the androgen pathway at multiple points: some medicines reduce testosterone production, while others block the androgen receptor that cancer cells use to receive growth signals. More complete disruption of this pathway may explain why combination therapy was associated with deeper PSA responses.

The findings could help physicians identify patients who may need closer monitoring or earlier treatment escalation. A patient whose PSA remains above 0.2 ng/mL during the first nine months of therapy may have a less favorable response and could be considered for additional treatment strategies, depending on overall health, cancer distribution, symptoms, genomic characteristics, and treatment goals. The researchers emphasize that the study shows an association rather than proving that changing therapy solely to force a lower PSA will extend life. Nevertheless, the results provide a practical target for evaluating response in routine care and reinforce the importance of achieving the deepest possible disease control in metastatic prostate cancer.

“ These data show that we need to target a PSA below 0.2 ng/ml as our metric of success to optimize outcomes for our patients with metastatic prostate cancer,” said corresponding author Stephen J. Freedland, MD, professor of urology at Cedars-Sinai and staff physician at the Durham VA Medical Center. Because the analysis was observational and based largely on Veterans Health Administration patients, future studies will need to determine whether the same threshold applies across broader populations and whether treatment decisions guided by this PSA target can improve survival in prospective clinical trials.

Subject of Research: Metastatic hormone-sensitive prostate cancer and PSA response to androgen-deprivation therapy

Article Title: How low do you need to go? Association between various prostate-specific antigen response measures and clinical outcomes in metastatic castration-sensitive prostate cancer in the Veterans Health Administration data

News Publication Date: August 10, 2026

Web References: CANCER journal: https://acsjournals.onlinelibrary.wiley.com/journal/10970142; DOI: https://doi.org/10.1002/cncr.70494

References: Stephen J. Freedland, Wei Gao, Maëlys Touya, Hongbo Yang, Jingyi Chen, Grace Chen, and Jasmina I. Ivanova. “How low do you need to go? Association between various prostate-specific antigen response measures and clinical outcomes in metastatic castration-sensitive prostate cancer in the Veterans Health Administration data.” CANCER. Published online August 10, 2026. DOI: 10.1002/cncr.70494

Keywords: Prostate cancer, metastatic cancer, PSA, prostate-specific antigen, androgen-deprivation therapy, hormone therapy, testosterone suppression, cancer treatment response, survival, oncology, urology, Veterans Health Administration

The study centers on mathematical biomarkers calculated from PSA measurements collected during a patient’s initial treatment cycle. Rather than treating PSA as a simple clinical signal that rises or falls, the researchers analyzed its underlying dynamics through mechanism-based mathematical models. These models are designed to separate different biological influences, such as the growth of treatment-sensitive cancer cells, the persistence or expansion of treatment-resistant populations, and the rate at which tumor-related PSA production changes over time. The resulting metrics offer a quantitative description of tumor behavior rather than merely a visual summary of the PSA trajectory.

This distinction is important because two patients can show superficially similar PSA patterns while harboring very different disease processes. A temporary decrease, for example, may represent a durable response in one patient but a short-lived suppression before resistant disease emerges in another. Traditional phenomenological models can fit the observed data, but they generally focus on reproducing the shape of the curve. Mechanism-based models instead attempt to infer the biological parameters that generated the curve, potentially making the measurements more useful for forecasting what happens next.

According to the study description, the investigators tested whether these model-derived biomarkers could predict patient-specific outcomes and survival. The metrics obtained from the first treatment cycle accurately identified differences among patients, suggesting that early PSA behavior contains more prognostic information than is captured by conventional monitoring approaches. The researchers reported that their mechanism-based biomarkers outperformed traditional phenomenological PSA measures in predicting clinically meaningful outcomes.

The technical foundation of the approach is mathematical parameter estimation. A patient’s PSA observations are fitted to equations representing competing or interacting tumor-cell populations and their treatment responses. The model then estimates quantities that cannot be observed directly in routine care, including effective growth rates, treatment sensitivity, and the relative contribution of disease compartments with different biological behaviors. These estimates can be converted into biomarkers that summarize the patient’s inferred cancer dynamics in a form suitable for statistical comparison and clinical prediction.

The potential advantage is speed. If reliable predictions can be made from the initial treatment cycle, clinicians may not need to wait for months of conventional monitoring before identifying a patient whose disease is unlikely to respond adequately. Early information could support closer surveillance, additional testing, or consideration of a different treatment strategy. Conversely, patients whose mathematical profiles indicate a favorable response might avoid unnecessary escalation. The study does not establish that model-guided treatment improves survival, but it provides evidence that the approach could become a decision-support tool for more individualized care.

The researchers describe the biomarkers as accessible because they are derived from PSA data already collected in routine prostate cancer management. This could make the framework easier to integrate into clinical workflows than approaches requiring new tissue sampling, specialized imaging, or complex molecular assays. However, mathematical accessibility does not eliminate the need for clinical validation. Before such biomarkers can guide treatment decisions, they would need to be tested prospectively across diverse patient populations, treatment settings, and measurement schedules.

The findings also illustrate a broader change in oncology: the shift from static biomarkers toward dynamic ones. A single measurement can indicate the state of a disease at one moment, while a time series can reveal how that disease reacts to pressure. Mathematical models provide a way to translate those changing signals into estimates of biological behavior. In prostate cancer, where treatment response and resistance can unfold over time, this dynamic perspective may be especially valuable.

Kit Gallagher, PhD, of the Department of Molecular Pathology at Mass General Brigham Cancer Institute, and Alexander R. Anderson, PhD, of the H. Lee Moffitt Cancer Center, are the study’s corresponding authors. Their work presents PSA not simply as a surveillance marker but as a source of mechanistic information. If further studies confirm the reported performance, model-based PSA biomarkers could help transform an inexpensive, widely available blood test into a mathematically informed system for stratifying patients and designing personalized treatment protocols.

Subject of Research: Mechanism-based mathematical biomarkers derived from PSA dynamics for predicting outcomes and survival in patients with prostate cancer.

News Publication Date: Not provided.

Web References: https://doi.org/10.1001/jamaoncol.2026.2781

References: Gallagher K, Anderson AR, et al. Study published in JAMA Oncology. DOI: 10.1001/jamaoncol.2026.2781.

Keywords: Prostate cancer, PSA dynamics, mathematical modeling, mechanism-based biomarkers, cancer biomarkers, treatment response, patient monitoring, survival prediction, personalized oncology, medical decision support

For decades, the medical community has operated under the assumption that elevated testosterone levels promote the growth and aggressiveness of prostate cancer. Testosterone, as a primary androgen hormone, has traditionally been implicated in fueling the proliferation of malignant prostate cells, influencing treatment protocols aimed at lowering androgen levels. However, the new retrospective cohort study published in The Journal of Urology presents a counterintuitive narrative, suggesting that men with low baseline testosterone levels—specifically those measuring 300 ng/dL or below—face a 60% higher likelihood of their prostate cancer advancing to a more aggressive Grade group 3 or higher during active surveillance.

Active surveillance is currently considered a safe and effective management strategy for patients diagnosed with low-risk, localized prostate cancer. It entails close monitoring of the disease, using repeated measurements of prostate-specific antigen (PSA), imaging scans, and targeted biopsies, delaying active treatment interventions unless there is evidence of disease progression. The complexity lies in accurately identifying which patients will maintain indolent disease versus those whose cancer will evolve into a more threatening form. The newly uncovered association between low testosterone and heightened progression risk could become a pivotal piece in this clinical puzzle.

The research team conducted a comprehensive analysis of clinical and pathological data from over 900 men undergoing active surveillance for localized prostate cancer. They meticulously controlled for potential confounding variables, including age, PSA levels, body mass index (BMI), tumor size, and density, to isolate the role of testosterone. The robustness of their findings indicates that testosterone levels at diagnosis might serve as an independent biomarker in predicting disease trajectory. This revelation prompts a reconsideration of hormonal influences in early-stage prostate cancer biology and suggests a more nuanced relationship than previously understood.

Dr. Justin R. Gregg, M.D., associate professor of Urology and Health Disparities Research at MD Anderson and lead author of the study, emphasizes the significance of these findings. He notes that recognizing the hormonal milieu’s impact on cancer progression can enhance personalized surveillance protocols. This could lead to stratifying patients not only based on traditional clinical parameters but by integrating endocrine profiles, creating a more refined and dynamic risk assessment model.

Biologically, the mechanisms underlying why low testosterone correlates with a more aggressive cancer course remain to be fully elucidated. Some hypotheses in the field propose that low androgen environments might select for more dedifferentiated, aggressive cancer clones that are less dependent on hormonal signals, potentially driving disease progression through alternative pathways. This is a stark contrast to the earlier view of testosterone purely as a growth facilitator, illuminating the complexity of endocrine interactions in prostate carcinogenesis.

It is critical to clarify that while this study identifies an association, it does not establish causality. Low testosterone per se is not deemed the cause of aggressive prostate cancer but rather a potential indicator or consequence of tumor biology that predisposes to disease progression. Future prospective studies are necessary to confirm whether baseline testosterone can be reliably used in clinical decision-making frameworks, guiding timing and frequency of surveillance biopsies and imaging, and determining when to transition to definitive treatment.

This research also opens questions about the role of testosterone replacement therapy (TRT) in men with prostate cancer or those at risk. Historically contraindicated due to fears of promoting tumor growth, the emerging evidence from this study and others might eventually support revisiting clinical guidelines. Nonetheless, caution remains paramount given the complexities of androgen signaling and prostate cancer pathophysiology.

For patients and clinicians alike, the study reinforces the importance of a comprehensive approach to prostate cancer management. Measuring testosterone at baseline could become standard practice, aiding in identifying men who might benefit from intensified surveillance or earlier intervention. As personalized medicine continues to evolve, integrating hormonal biomarkers with genomic and imaging data could revolutionize prostate cancer care, minimizing overtreatment while safeguarding against missed progression.

In conclusion, this major study from MD Anderson centers a critical spotlight on the relationship between testosterone and prostate cancer behavior during active surveillance. It welcomes a new era of research dedicated to unraveling endocrine factors in cancer progression, which, if confirmed by subsequent investigations, holds the promise of transforming prostate cancer prognostication and therapeutic decision-making. The findings underscore the necessity of embracing a multidimensional view of cancer biology that transcends traditional dogmas and paves the way for informed, patient-centered care.

Subject of Research: The relationship between baseline testosterone levels and prostate cancer progression in patients under active surveillance.

Article Title: Low Testosterone Levels and Grade Group Progression Among Localized Prostate Cancer Patients on Active Surveillance: A Retrospective Cohort Study

News Publication Date: 24-Feb-2026

Web References:

• The University of Texas MD Anderson Cancer Center
• The Journal of Urology article DOI: 10.1097/JU.0000000000004986

References: Gregg, J. R., et al. (2026). Low Testosterone Levels and Grade Group Progression Among Localized Prostate Cancer Patients on Active Surveillance: A Retrospective Cohort Study. The Journal of Urology. DOI: 10.1097/JU.0000000000004986

Keywords: Prostate cancer, testosterone, active surveillance, cancer progression, hormone biomarkers, Grade Group progression, endocrine factors, tumor biology, personalized medicine