© The Author 2007. Published by Oxford University Press.
CORRESPONDENCE |
Response: Re: Rising Incidence of Small Renal Masses: A Need to Reassess Treatment Effect
Affiliations of authors: Department of Urology, University of Michigan, Ann Arbor, MI (JMH, SD, BKH); Department of Urology, University of California, Los Angeles, CA (DCM)
Correspondence to: Brent K. Hollenbeck, MD, MS, Department of Urology, University of Michigan, 1500 East Medical Center Dr, 3876 Taubman Center, Ann Arbor, MI 48109 (e-mail: bhollen{at}med.umich.edu).
Chow and colleagues raise three reasonable concerns regarding our recent manuscript.
The first concern relates to the inclusion of certain International Classification of Diseases for Oncology, 2nd Edition (ICD-O-2) histology codes—8960 (Wilms tumor), 8963 (rhabdoid sarcoma), and 8041 (small-cell carcinoma)—as part of the analysis. Admittedly, these rare histologies are not classically grouped with renal cell carcinomas; however, they accounted for only 2.36% (n = 713) of all incident cases over the study interval. In addition, their exclusion affected our models very little. The parameter estimates for tumor size and follow-up time remain nearly identical. The slight changes seen in the age parameter estimates would be expected with the exclusion of the Wilms tumor cases, which are largely pediatric malignancies. However, these small changes do not alter the magnitude or direction of the incidence, treatment, and mortality trends for kidney cancer. Hence, the overall message—increased rates of treatment accompanied by increased mortality rates—remains unchanged. They also made a case for inclusion of the ICD-O-2 code for papillary adenocarcinoma (8340), but there were no cases with this code in our original sample even before exclusions were made based on tumor size information.
A second concern derives from our use of the data from the "restricted" period (1983–2002) for our analysis, as opposed to from the full period for which some Surveillance, Epidemiology, and End Results (SEER) data were available (1973–2002). As we state in our introduction, our intent was to "derive incidence, treatment, and mortality trends for kidney cancer ... as a function of tumor size." Tumor size information for incident cases of kidney cancer ascertained by the SEER registries was not collected until 1983. Therefore, in our analysis, the years 1973–1982 were excluded. It is important to note that, although the inclusion of all years’ data, as suggested by Chow et al., attenuates the mortality trend, mortality rates have still risen considerably.
A final concern centered on our choice of endpoint. Chow et al. argue that survival rates, rather than mortality rates, "more accurately reflect the clinical course of patients following cancer diagnosis." We agree that 5-year survival is a valid measure of comparison for various cancer therapies in the setting of a clinical trial. However, such comparisons across time or place may be misleading. Given the problems related to detection bias as diagnostic thresholds change, some have argued that progress against cancers be assessed using population-based mortality rates (1,2). Using data from SEER, Welch et al. (3) measured the relationship over time between 5-year cancer-specific survival, mortality, and incidence for 20 solid tumor types (including kidney cancer). They found that increases in 5-year survival over time had little relationship to changes in the mortality from cancer. Rather, improvements in survival appeared to be related to changing patterns of diagnosis.
We thank Drs Chow, Linehan, and Devesa for their interest in our work and appreciate their insight. As detection thresholds for disease entities change with advances in medical technology, clinicians need to have a better understanding of their evolution to provide the best counseling to their patients. Furthermore, we should remain mindful that unnecessary treatment always has a negative impact on the quality of health care.
REFERENCES
(1) Bailar JC, Smith EM. Progress against cancer? N Engl J Med (1986) 314:1226–32.[Abstract]
(2) Extramural Committee to Assess Measures of Progress Against Cancer. Measurement of progress against cancer. J Natl Cancer Inst (1990) 82:825–35.
(3) Welch HG, Schwartz LM, Woloshin S. Are increasing 5-year survival rates evidence of success against cancer? JAMA (2000) 283:2975–8.
Related Correspondence
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2007 99: 569-570.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||