This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Komaki, R. Articles by Allen, P. K. Search for Related Content PubMed Articles by Komaki, R. Articles by Allen, P. K. Related Articles Related Correspondence Social Bookmarking                            What's this?

In Reply:

The University of Texas M.D. Anderson Cancer Center, Houston, TX

We thank Dr Atkins for his comments. Dr Atkins suggested that the prognostic significance of the standardized uptake value (SUV) in non–small-cell lung cancer,¹ might be overestimated with reference to a report by Dr Altman et al.² We agree with Dr Atkins that any investigator needs to pay full attention to avoiding false-positive results. However, we disagree with his suggestion for the following reasons. First, Dr Atkins recognized SUV as a very simple continuous variable such as the S phase fraction (SPF) in breast cancer and said our data should be corrected by the formula proposed by Dr Altman et al. However, the SUV should be carefully considered and determined by many factors, such as the dose of [¹⁸F]fluorodeoxyglucose (FDG), the time needed to acquire images and data, determination of the region of interest, methods of reconstruction, fasting, and the blood sugar level of the patient. Therefore, the SUV is a continuous variable; however, we could not confirm that the SUV should be treated like the SPF. For this same reason, a comparison of SUVs between institutions is quite difficult.

Second, Dr Altman et al² recommended in their conclusion that to avoid false-positive results, it is better to use three or four subgroups rather than two, and if an investigator needs to analyze two subgroups, the investigator should use the correcting formula of Dr Altman et al. Using their recommendation, we reanalyzed the prognostic importance of the SUV of the primary tumor with three subgroups (SUV 4, n = 32; 4 < SUV 10, n = 70; 10 < SUV, n = 60). The prognostic importance of the SUV was conserved for both overall survival (P = .027 by log-rank test, P = .029 by Wilcoxon test) and disease-free survival (P = .0042 by log-rank test, P = .0021 by Wilcoxon test). Therefore, we believe that the SUV of the primary tumor was not overestimated and is still a significant prognostic factor and a useful decision-making tool.

Third, as shown in our previous report,¹ the SUV was determined to be an independent and significant prognostic factor by multiple regression analysis using the Cox proportional hazards model in both overall and disease-free survival.

Moreover, a report combining four Radiation Therapy Oncology Group studies showed that multiple factors, including weight loss, Karnofsky performance status, node stage, and age, affect outcomes in inoperable non–small-cell lung cancer.³ It is important to note that the SUV of the primary tumor in our series showed stronger prognostic value than any of those factors.

As Dr Atkins suggested, a larger investigation is needed and, in fact, is now ongoing. ACRIN 66,668/RTOG 0235 has been activated, and we will accrue a large number of patients to determine the optimal cutoff point of the SUV of the primary tumor for future patients with unresectable non–small-cell lung cancer.

Authors' Disclosures of Potential Conflicts of Interest

The authors indicated no potential conflicts of interest.

REFERENCES

1. Sasaki R, Komaki R, Macapinlac H, et al: [18F]Fluorodeoxyglucose uptake by positron emission tomography predicts outcome of non–small-cell lung cancer. J Clin Oncol 23:1136-1143, 2005[Abstract/Free Full Text]

2. Altman DG, Lausen B, Sauerbrei W, et al: Dangers of using "optimal" cutpoints in the evaluation of prognostic factors. J Natl Cancer Inst 86:829-835, 1994[Free Full Text]

3. Komaki R, Scott CB, Byhardt R, et al: Failure patterns by prognostic group determined by recursive partitioning analysis (RPA) of 1547 patients on four radiation therapy oncology group (RTOG) studies in inoperable nonsmall-cell lung cancer (NSCLC). Int J Radiat Oncol Biol Phys 42:263-267, 1998[CrossRef][Medline]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

Related Correspondence

• Overestimation of the Prognostic Significance of SUV Measurement by Positron Emission Tomography for Non–Small-Cell Lung Cancer
  Carl D. Atkins
  JCO 2005 23: 6799-6800 [Full Text]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Komaki, R. Articles by Allen, P. K. Search for Related Content PubMed Articles by Komaki, R. Articles by Allen, P. K. Related Articles Related Correspondence Social Bookmarking                            What's this?