|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2005.07.909 on September 26 2005 © 2005 American Society of Clinical Oncology.
Rebalancing Ratios and Improving Impressions: Later Thoughts From the Prostate Cancer Prevention Trial InvestigatorsCenter for Outcomes Research, Massachusetts General Hospital Cancer Center, Massachusetts General Hospital, and Harvard Medical School, Boston, MA
By defining type I and type II errors of inference, the great biostatisticians Neyman and Pearson gave medical investigators both a powerful analytic tool and a crucial perspective for evaluating medical innovations. A type I error rejects the null hypothesis when it is true, whereas a type II error accepts it when it is not. By keeping these categories in mind, Neyman and Pearson correctly reasoned, investigators could explicitly weigh the consequences of mistaken conclusions in opposite directions. Depending on circumstances, one error or the other would likely cause greater harm. The cost of a type I error increases when the treated condition is less harmful; the study intervention is more toxic, invasive, or costly; and attractive alternatives are available. Conversely, type II errors weigh more heavily when the condition is dire, the intervention benign and inexpensive, and alternatives are unavailable or unappealing. After considering the clinical situation, investigators could tilt the study design or analysis plan away from the more damaging error, at the cost of increased risk of the more benign. Requiring stronger evidence that the null hypothesis is unlikely, by reducing the Although Neyman and Pearson’s approach applies most directly to decisions to reject or accept the null hypothesis, it is broadly relevant to choices made under uncertainty: if we make a mistake, it should be the least damaging. In their examination of factors that might alter the previously reported, conflicting results of the Prostate Cancer Prevention Trial (PCPT) in the current issue of the Journal of Clinical Oncology, Klein et al1 give Neyman and Pearson’s insight only glancing attention. Rather than examining the relative consequences of the opposing effects found in their path-breaking study, they instead focus on assumptions that would make using finasteride to prevent prostate cancer look more attractive than their originally published results suggested.2
To determine whether finasteride could prevent prostate cancer, the designers of the PCPT required a persuasive study outcome. Because of prostate cancer’s long natural history, the definitive and unambiguous outcome, mortality, which would require 15 or more years of follow-up, was out of practical reach. The study outcome they chose, the diagnosis of prostate cancer, posed difficulties because the tumor is so prevalent: it can be found in the prostates of more than half of men older than age 50. As a result, an encounter with a urologist and, subsequently, his biopsy gun is the most powerful risk factor for its diagnosis. This observation seems jokey, a tautologic observation that to diagnose this cancer—like many others—requires a biopsy and a surgeon to perform it. But those familiar with prostate cancer’s astonishing recent history and its intimate association with prostate-specific antigen (PSA) screening, which certainly include the PCPT investigators, understand its clinical importance. Practice patterns have changed the incidence of prostate cancer dramatically and suddenly. For example, in the month after the uncontrolled observation was prominently published that systematic prostate biopsies for a PSA level Given that the goal of finasteride therapy in the PCPT was not to inhibit the diagnosis of prostate cancer but rather to reduce its presence in the prostate gland, the usual for-cause biopsy criterion—an elevated PSA level—was unreliable, given that finasteride suppresses it. Therefore, the study’s designers decided that any patient leaving the study without a for-cause biopsy would get one: every prostate would be sampled. The biopsy all strategy neutralized the potential criticism that vagaries in biopsy rates, not true changes in cancer progression, might explain different observed diagnosis rates, but it also produced a startling demonstration of the vast potential for finding prostate cancer when asymptomatic men undergo biopsy willy-nilly. When the study was halted, one in four patients in the control arm had been diagnosed with prostate cancer within 7 years, with another 10% still not biopsied.2 Given the 3% lifetime risk of death from prostate cancer, the vast majority of cancers diagnosed by blanket biopsy were no threat to the men in whom they were diagnosed. However, the result that prompted the present analysis of Klein et al1 was an even more unexpected outcome: judged by Gleason score, the best predictor of a prostate cancer’s aggressiveness, the effect of finasteride varied by the biologic potential of the cancer. Diagnoses of nonaggressive cancers (Gleason score 6 or less) were decreased 25% in the finasteride arm, from 24.4% to 18.4%, or 6.0%, but approximately 27% more high-grade cancers were found (Gleason score 8 to 10, grouped together with Gleason score 7 cancers, which were unchanged), indicating an absolute increase of 1.3%. Using the absolute values, Klein et al present these divergent results in a ratio, with the favorable outcome (the reduction in nonaggressive cancers) in the numerator and the unfavorable outcome (the increase in high-grade cancers) in the denominator. Expressed this way, higher ratios make finasteride use more attractive and lower values less so. From this starting point, the investigators explored theoretical corrections for two postulated artifacts that might have affected the ratio unfavorably. The first affected the numerator: by reducing benign prostate tissue, finasteride may have relatively increased the proportion of the prostate tissue a biopsy samples, artificially inflating the intervention arm’s prostate cancer detection rate. The second affected the denominator: the authors propose that finasteride may have produced tissue artifacts leading to overdiagnosis of high-grade cancers. Applying progressively stronger assumptions about the magnitude of these effects increased the benefit/risk ratio from 4.6:1 to 82:1. In addition, they considered another assumption that might reduce the ratio, including only cancers diagnosed with for-cause biopsies, producing ratios from a baseline of 1.9 to a corrected ratio of 8.0. However, they left the results out of the abstract and gave them short shrift in the Discussion. Perhaps more important, they spent no effort considering the qualitative differences between these opposing effects and their relative importance. A simple numeric ratio implicitly gives a decrease in low-grade cancers the same weight as an increase in high-grade cancers. When the authors state "grade 6 disease is certainly relevant clinically," they are correct, but the damage from nonaggressive cancers, unlike high-grade cancers, is iatrogenic, not biologic. In this country, prostate cancers are almost invariably treated, subjecting the patients to treatment-related complications, which include the probability of long-term, severe erectile dysfunction and the possibility, depending on treatment modality, of urinary incontinence or bowel problems. Treatment policies that reduce the proportion of nonaggressive cancers undergoing treatment, or reduce the likelihood of biopsy and their diagnosis in the first place, would also reduce this damage. The biology of high-grade cancers, which disproportionately produces poor outcomes, is less malleable by changes in practice patterns. From the patient’s perspective, increasing high-grade cancers and lowering low-grade cancers by the same amount would not be a wash. In short, the authors’ analysis shows little interest in exploring the clinical implications of the ratio they created or possibilities that might make it less favorable. Rather, the analysis is biased toward analyses that make the clinical outcomes, and the finasteride intervention, more appealing.
The authors do not discuss a mechanism that might explain the conflicting results, both good and bad. Let me try here. The PCPT provided scientific (and perhaps politically appealing) symmetry with a large chemoprevention study initiated earlier in breast cancer.5 The agents in the two studies differ but may have an interesting parallel. The breast cancer study agent tamoxifen, like other selective estrogen receptor modulators used in later breast cancer chemoprevention studies, binds estrogen receptors (ERs), physically excluding natural estrogens. Whether the interaction activates the receptors or not depends on the tissue-specific ER and physical properties of the molecule that binds it. Furthermore, the interaction between tamoxifen and the ER in breast cancer may change over time; benefit from adjuvant tamoxifen therapy for more than 5 years is unproven and may be harmful,6 and continuing it longer outside of a clinical trial is not recommended.7 The PCPT agent, the 5- The PCPT’s results are important—we just do not understand them. Klein et al1 have done a service by examining them in a different light, although our gratitude would be greater had they been more even-handed. The study’s most pressing legacy is the challenge for others to try. Author's Disclosures of Potential Conflicts of Interest The author indicated no potential conflicts of interest. REFERENCES
1. Klein EA, Tangen CM, Goodman PJ, et al: Assessing benefit and risk in the prevention of prostate cancer: The Prostate Cancer Prevention Trial revisited. J Clin Oncol 23:7460-7466, 2005
2. Thompson IM, Goodman PJ, Tangen CM, et al: The influence of finasteride on the development of prostate cancer. N Engl J Med 349:215-224, 2003 3. Catalona WJ, Smith DS, Ratliff TL, et al: Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N Engl J Med 324:1156-1161, 1991[Abstract] 4. Giovannucci E, Kantoff P, Spiegelman D, et al: The epidemic of prostate cancer and the medical literature: A causal association? Prostate Cancer Prostatic Dis 1:148-153, 1998[CrossRef][Medline]
5. Fisher B, Costantino JP, Wickerham DL, et al: Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 90:1371-1388, 1998 6. Tamoxifen for early breast cancer: An overview of the randomised trials—Early Breast Cancer Trialists' Collaborative Group. Lancet 351:1451-1467, 1998[CrossRef][Medline]
7. Goldhirsch A, Wood WC, Gelber RD, et al: Meeting highlights: Updated international expert consensus on the primary therapy of early breast cancer. J Clin Oncol 21:3357-3365, 2003
8. Rittmaster RS: Drug therapy: Finasteride. N Engl J Med 330:120-125, 1994 9. Gormley GJ, Stoner E, Bruskewitz RC, et al: The effect of finasteride in men with benign prostatic hyperplasia: The Finasteride Study Group. N Engl J Med 327:1185-1191, 1992[Abstract]
10. Taplin ME, Bubley GJ, Shuster TD, et al: Mutation of the androgen-receptor gene in metastatic androgen-independent prostate cancer. N Engl J Med 332:1393-1398, 1995 11. Chen CD, Welsbie DS, Tran C, et al: Molecular determinants of resistance to antiandrogen therapy. Nat Med 10:33-39, 2004[CrossRef][Medline]
Related Article
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
value or increasing the size of the clinically meaningful benefit the study is powered to detect, makes a type I error less likely, whereas greater study power, from increased enrollment or longer follow-up, reduces the chance of a type II error by producing a more precise estimate of the treatment effect. 
4.0 ng/mL frequently detect prostate cancers in asymptomatic men,
