Behind the Paper

The importance of confirming mCSPC classifications in real-world evidence studies

Stephen Freedland Sep 22, 2025

Precision in clinical definitions is critical to the credibility of real-world evidence studies. Among patients with prostate cancer (PC)—the most common cancer in men and the second leading cause of cancer death in the United States1— approximately 5–10% present with metastatic disease at the time of diagnosis.2,3 These patients have an aggressive disease course and have a worse prognosis than patients diagnosed with localized disease.4  Among patients with metastatic disease, classifying disease stage into metastatic castration-sensitive prostate cancer (mCSPC) or metastatic castration-resistant prostate cancer (mCRPC) is essential as patients with mCRPC require different treatment approaches and, importantly, have much worse clinical outcomes and overall survival than patients with mCSPC. However, misclassification between mCSPC and mCRPC remains a persistent challenge that can have significant impacts on the validity and interpretation of real-world evidence studies, especially those relying on administrative claims or electronic health records data.

Our study, recently published in Prostate Cancer and Prostatic Diseases, evaluated the real-world effectiveness of treatment intensification with androgen receptor pathway inhibitors (ARPIs) in patients with mCSPC.5 A key first step in this evaluation was to ensure we were studying the right patients. Defining this population with accuracy, however, is a complex task.

Why it matters

Within the International Classification of Diseases (ICD), no codes exist to distinguish between stages of PC. Although codes exist to denote metastasis, and are widely used, codes to indicate the presence of castration resistance are seldom used in clinical practice.6,7 This is particularly challenging as disease states can change over time, adding to the difficulty of classifying patients at any given time point. As a result, many real-world evidence studies (including several of our own previous studies8-10) rely on case-finding algorithms based on factors such as health claims data and laboratory test results. These algorithms must be developed and validated carefully to avoid inadvertently including both patients with mCSPC and patients with mCRPC, an error that can lead to misleading study conclusions, especially when evaluating drugs that are approved in both mCSPC and mCRPC.

Our approach: Electronic medical record selection with chart review validation

To avoid this pitfall, we took a detailed approach to patient eligibility and selection. De novo mCSPC patients were selected using an algorithm evaluating PSA rise and treatments using data from the Veterans Affairs (VA) electronic health record system and thorough chart review was conducted to confirm their diagnoses. Specifically, adult patients were only included if their mCSPC diagnosis was confirmed by this clinical chart review (see manuscript for detailed enrollment criteria). Within mCSPC patients, we selected patients with a de novo mCSPC diagnosis between February 1, 2018 and June 30, 2020 who had received one of the following treatments: first-line androgen-deprivation therapy (ADT) alone, ADT + non-steroidal anti-androgens (NSAA), or ADT + ARPI ± NSAA. Patients who had previously been treated with cabazitaxel, docetaxel, mitoxantrone, carboplatin, cisplatin, oxaliplatin, radium-223, or sipuleucel-T any time before the index date (i.e., date of ADT initiation) were excluded. These treatment approaches are frequently used among patients with mCRPC; thus, by excluding these patients, we reduced the risk of inadvertently mixing disease states in our analysis, strengthening the validity of our findings. To further reduce the risk of bias in our analysis, we also used inverse probability of treatment weighting (IPTW) to help balance baseline characteristics between patients within each of the three treatment groups.

What we found

Among 384 patients with confirmed de novo mCSPC, median follow-up time was 37.2, 38.1, and 34.8 months for ADT alone, ADT + NSAA, and ADT + ARPI, respectively. Across our IPTW-adjusted analyses, our results showed that patients treated with ADT + ARPI had significantly better outcomes than those treated with ADT alone. Specifically, compared with ADT alone, patients treated with ADT + ARPI had:

  • A 39% lower risk of death
  • A 60% lower risk of progression to mCRPC
  • Greater prostate-specific antigen (PSA) declines across all thresholds and were more likely to achieve PSA reduction ≥50%, despite a higher median baseline PSA level (162.7 ng/mL vs 121.6 ng/mL for ADT alone)

These findings for overall survival (IPTW-adjusted hazard ratio: 0.61 [95% confidence interval: 0.44 to 0.87]) mirror those from pivotal clinical trials like STAMPEDE (0.60 [0.48−0.73])11, LATITUDE (0.62 [0.51 to 0.76])12, TITAN (0.65 [0.53 to 0.79])13, ARCHES (0.66 [0.53 to 0.81])14, and ENZAMET (0.60 [0.47 to 0.78])15, suggesting that the survival benefits of ARPIs extend into real-world practice when the right patients are identified.

The bigger message

As more therapies become available for the treatment of mCSPC, real-world evidence studies will play an important role in evaluating these therapies in clinical practice. Yet, without rigorous diagnostic precision, these evaluations risk providing clinicians and patients with inaccurate guidance, particularly for therapies that are approved for both mCSPC and mCRPC. While algorithms that combine diagnostic codes, claims data, and laboratory data offer an important starting point for real-world evidence studies, chart review can help to ensure that complex disease states are being accurately defined. Indeed, in our study, chart review led to the same conclusions drawn in prior algorithm-based studies. Thus, although chart review is a time-consuming practice, incorporating its use into real-world evidence studies can help to validate algorithms and strengthen our confidence in their conclusions.

In short: if we want real-world evidence to be truly meaningful, we must make sure our classifications are confirmed and validated. As more decisions are made based on real-world data, chart review offers an important resource for identifying the right population, validating findings from studies that rely on claims-based algorithms, and helping to ensure that we draw the correct conclusions.

  

 References:

  1. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin. Jan-Feb 2024;74(1):12-49. doi:10.3322/caac.21820
  2. Centers for Disease Control and Prevention. Prostate Cancer Incidence by Stage at Diagnosis. Accessed August 20, 2025. https://www.cdc.gov/united-states-cancer-statistics/publications/prostate-cancer.html#cdc_research_or_data_summary_explore_more-data-source
  3. Kelly SP, Anderson WF, Rosenberg PS, Cook MB. Past, Current, and Future Incidence Rates and Burden of Metastatic Prostate Cancer in the United States. European Urology Focus. 2018/01/01/ 2018;4(1):121-127. doi:https://doi.org/10.1016/j.euf.2017.10.014
  4. Finianos A, Gupta K, Clark B, Simmens SJ, Aragon-Ching JB. Characterization of Differences Between Prostate Cancer Patients Presenting With De Novo Versus Primary Progressive Metastatic Disease. Clin Genitourin Cancer. Aug 31 2017;doi:10.1016/j.clgc.2017.08.006
  5. Freedland SJ, Hong A, El-Chaar N, et al. Survival benefit associated with first-line androgen receptor pathway inhibitors for de novo metastatic castration-sensitive prostate cancer. Prostate Cancer Prostatic Dis. Jul 25 2025;doi:10.1038/s41391-025-01000-8
  6. Freedland SJ, Ke X, Lafeuille M-H, et al. Identification of patients with metastatic castration-sensitive or metastatic castration-resistant prostate cancer using administrative health claims and laboratory data. Current Medical Research and Opinion. 2021/04/03 2021;37(4):609-622. doi:10.1080/03007995.2021.1879753
  7. Candelieri-Surette D, Hung A, Lynch JA, et al. Development and Validation of a Tool to Identify Patients Diagnosed With Castration-Resistant Prostate Cancer. JCO Clin Cancer Inform. Sep 2023;7:e2300085. doi:10.1200/cci.23.00085
  8. Freedland SJ, Davis M, Epstein AJ, Arondekar B, Ivanova JI. Real-world treatment patterns and overall survival among men with Metastatic Castration-Resistant Prostate Cancer (mCRPC) in the US Medicare population. Prostate Cancer and Prostatic Diseases. 2024/06/01 2024;27(2):327-333. doi:10.1038/s41391-023-00725-8
  9. Freedland SJ, Ke X, Lafeuille MH, et al. Identification of patients with metastatic castration-sensitive or metastatic castration-resistant prostate cancer using administrative health claims and laboratory data. Curr Med Res Opin. Apr 2021;37(4):609-622. doi:10.1080/03007995.2021.1879753
  10. Freedland SJ, Davis MR, Epstein AJ, Arondekar B, Ivanova JI. Healthcare Costs in Men with Metastatic Castration-Resistant Prostate Cancer: An Analysis of US Medicare Fee-For-Service Claims. Adv Ther. Oct 2023;40(10):4480-4492. doi:10.1007/s12325-023-02572-4
  11. Attard G, Murphy L, Clarke NW, et al. Abiraterone acetate and prednisolone with or without enzalutamide for high-risk non-metastatic prostate cancer: a meta-analysis of primary results from two randomised controlled phase 3 trials of the STAMPEDE platform protocol. Lancet. Jan 29 2022;399(10323):447-460. doi:10.1016/S0140-6736(21)02437-5
  12. Fizazi K, Tran N, Fein L, et al. Abiraterone plus Prednisone in Metastatic, Castration-Sensitive Prostate Cancer. N Engl J Med. Jul 27 2017;377(4):352-360. doi:10.1056/NEJMoa1704174
  13. Chi KN, Chowdhury S, Bjartell A, et al. Apalutamide in Patients With Metastatic Castration-Sensitive Prostate Cancer: Final Survival Analysis of the Randomized, Double-Blind, Phase III TITAN Study. J Clin Oncol. Jul 10 2021;39(20):2294-2303. doi:10.1200/JCO.20.03488
  14. Armstrong AJ, Azad AA, Iguchi T, et al. Improved Survival With Enzalutamide in Patients With Metastatic Hormone-Sensitive Prostate Cancer. J Clin Oncol. May 20 2022;40(15):1616-1622. doi:10.1200/JCO.22.00193
  15. Davis ID, Martin AJ, Stockler MR, et al. Enzalutamide with Standard First-Line Therapy in Metastatic Prostate Cancer. N Engl J Med. Jul 11 2019;381(2):121-131. doi:10.1056/NEJMoa1903835