RSS Feeds
-
In Search of Valid Prognostic Markers
Submit responseIn reply to our editorial Concato et al argue three major points. Firstly they say that their study determined whether marker status at diagnosis (versus post-surgery) predicted actual mortality. They emphasize the importance of not relying on biochemical failure is an equivalent outcome to mortality. Secondly they explain that their results can guide future research on biologic mechanisms and suggest new approaches to therapy. Thirdly they cite other reviews as affirming the importance of the three markers they highlighted, P53, BCL-2, and microvessel density. We commended Concato et al for assembly of a large and important study cohort. However in trying to identify important prognostic markers Concato et al did not surpass the power of Gleason scoring nor did they compare directly their data with histologic grade in their search for clinically useful prognostic indicators. Lastly, Concato et al did not address the possibility that these markers are predictive of response to various therapies as well as prognostic. Thus the use of irradiation, hormonal, or chemotherapy should have been taken into account in the analysis (1). We also draw attention to the guidelines published by McShane et al that set criteria for studies of prognostic markers (2). Although the data presented by Concato are not in compliance with a number of these guidelines, we were most concerned that the assay methods were neither properly validated nor set forth in a manner that allowed replication and confirmation by others. Moreover, we noted that there was insufficient statistical power to demonstrate the effect of BCL-2 expression that appeared to be the most promising marker. Lastly, the choice of markers was not well justified either by the literature or by the potential of the markers to be of practical application in clinical practice. Further insight into the biological processes underlying aggressive disease is limited by the focus on known prostate cancer genes examined independently in published studies. We reaffirm our conclusions that P53 and microvessel density will not find application in clinical practice and the utility of BCL-2 awaits validation by studies that are adequately powered and include methodology consistent with published guidelines.
References
1. Henry NL, Hayes DF. Uses and abuses of tumor markers in the diagnosis, monitoring, and treatment of primary and metastatic breast cancer. Oncologist. 2006;11:541-52.
2. McShane LM, Altman DG, Sauerbrei W, Taube SE, Gion M, Clark GM. Reporting recommendations for tumor marker prognostic studies (REMARK). J Natl Cancer Inst. 2005;97:1180-1184.
Conflict of Interest:
We are the authors of the editorial.
-
Relevance in prostate cancer
Submit responseTo the Editor,
The editorial (1) on molecular markers in prostate cancer (2) may have generated several misconceptions. Our emphasis was on novel associations between molecular markers obtained at biopsy and long-term mortality. Although many articles “evaluating correlations of p53 and bcl- 2 and microvessel density” (1) had been published previously, a gap in knowledge still existed—whether marker status at diagnosis (versus post- surgery) predicts mortality (versus presuming biochemical failure is an equivalent outcome). By conducting a comprehensive review of medical records, retrieving original biopsy tissue, and using rigorous methods of clinical epidemiology, we demonstrated in a large, community-based population that these markers at diagnosis confer an independent risk of death from prostate cancer.
Yet, according to the editorial: the findings may not “have any effect at all on clinical practice”; and only “narrow-needle biopsy specimens” were examined using immunohistochemical techniques “despite the availability of modern high-throughput technologies”; and another “shortcoming of the study was the choice of molecular markers.”
Of note, we didn’t advocate widespread clinical use of these markers. As mentioned (2), our results can guide future research on biologic mechanisms and new approaches to therapy. The findings can also encourage a renewed focus on molecular markers. In terms of contemporary care, if few men will die from their prostate cancer, and if bcl-2 expression (for example) has low prevalence but predicts mortality, then a patient and his clinician might want to know such information. In this context, markers may not affect primary treatment for men when diagnosed with prostate cancer, but they can help to identify patients who will not respond and require treatment subsequently (perhaps involving trials of targeted therapies).
Importantly, we adopted a patient-centered approach not recognized in the editorial. Treatment decisions are made based on needle biopsies; and modern genome-wide studies (endorsed in the editorial) focus on etiology, not prognosis, while explaining only a small proportion of genetic risk (3). In addition, criticism of the markers we selected was arguably misguided. Other groups, and recent reviews (4, 5), consider these markers as among the most relevant prognostic factors.
Finally, this situation highlights challenges of maintaining a balanced perspective in academic publications. Including the word “incremental” to avoid overstating the implications of our findings was used against us in the editorial. So be it. We will let the medical community evaluate our work—and the editorial—as we address a major men’s health issue.
John Concato, M.D., M.P.H. Edward Uchio, M.D. Carolyn K. Wells, M.P.H. Veterans Affairs Clinical Epidemiology Research Center West Haven, CT 06516
References
1. Gelmann EP, Henshall SM. Clinically relevant prognostic markers for prostate cancer: the search goes on. Ann Intern Med. 2009;150:647-9.
2. Concato J, Jain D, Uchio E, Risch H, Li WW, Wells CK. Molecular markers and death from prostate cancer. Ann Intern Med. 2009;150:595-603.
3. Kraft P, Hunter DJ. Genetic risk prediction—are we there yet? New Engl J Med. 2009;360:1701-3.
4. McKenzie S, Kypriano N. Apoptosis evasion: the role of survival pathways in prostate cancer progression and therapeutic resistance. J Cell Biochem. 2006;97:18-32.
5. Schlomm T, Erbersdobler A, Mirlacher M, Sauter G. Molecular staging of prostate cancer in the year 2007. World J Urol. 2007;25:19-30.
Conflict of Interest:
We published the article that the editorial refers to.