Reply to radiotherapy for human papillomavirus‐positive oropharyngeal cancers in the National Cancer Data Base
Year: 2016
Type: letter
Abstract: We appreciate the interest in our article1 by Cheraghlou et al, and the opportunity provided by the editors for a response. It is important to note that the goal of our article and its conclusions was not to endorse radiation alone as therapy for patients with the most favorable presentations of human papillomavirus-related oropharyngeal cancer, but rather to attempt to validate whether the progression-free survival target goals set in cooperative trials for treatment deintensification were feasible, as well as in a large data set to determine whether we could identify subgroups for whom the deintensification of therapy may or may not be appropriate. The accompanying letter, using a National Cancer Data Base (NCDB) data set, suggests a more cautionary tone for radiotherapy alone as a form of deintensification for this “low-risk” group. Analysis of NCDB data has strengths and weaknesses. Proponents cite the opportunities afforded by these data to harness “big data” and evaluate “real-world” outcomes. Advocates of these large data sets believe that large numbers can compensate for data of lesser quality or with missing variables that may better explain the observations. Regardless, the data presented by Cheraghlou et al fail with regard to having large numbers; because the radiotherapy cohort comprised only 51 patients, <7% of the sample, and with an approximate 80% survival rate, the conclusions are based on only a few events. Comparatively, our cohort of patients treated with radiotherapy consisted of >300 patients, and had better balance compared with the chemotherapy arm (>400 patients).1 With regard to the real-world situation, the NCDB data are valuable in that in an era in which guidelines advocate chemoradiation for the majority of patients with this disease, >90% of patients in the NCDB were treated using these guidelines. However, as is common in NCDB articles, although demonstrating what is often done in the real-world, hypotheses concerning why these patterns of care are observed were less forthcoming. Despite controls for age and comorbidities, we speculate that those individuals chosen for radiotherapy alone in the NCDB more likely had poorer performance status, contributing to the poorer survival outcome observed. We can only speculate because the NCDB only provides survival data, and patterns of disease recurrence (or lack thereof) are absent. To the contrary, even within the narrow definition of a low-risk group, our data reflect that those chosen for lesser therapy were often chosen based on their having a lower tumor burden.1 Only approximately 20% of those with T1 disease were treated with chemoradiation compared with 66% of those with T3 disease. Furthermore, we cautioned that individuals with T3 disease had worse outcome. We would like to make 3 additional points. First, the data from our study described a phenotype that included patients with limited or never-smoking status.1 Numerous articles2 have demonstrated the negative association between smoking and survival and, again, smoking is not assessed in the current NCDB data analyses, and their cohort likely includes smokers. Second, the NCDB can only compare chemoradiation and radiation in abstract terms. Important details regarding both the drugs and radiation delivery that may explain outcomes are not available. Last, Cheraghlou et al may be correct that our findings may not be generalizable. The patients at our institution are treated at a high-volume academic center with a large team dedicated to the care of patients with head and neck cancer. Recent articles from the NCDB have demonstrated that both treatment at academic centers and treatment by teams with high levels of expertise are associated with significantly higher rates of survival.3, 4 Ultimately, both our study1 and that of Cheraghlou et al are retrospective analyses, and therefore are given to all the flaws inherent in such studies. In particular, the selection of therapy is clearly biased, and the differences in this selection between the patients in our study1 and those of the NCDB may explain these differences, although we can only speculate. To best answer the question, the normal process of years of cooperative trials followed by years of large data collection may give us the answers regarding dose deintensification for patients with human papillomavirus-associated oropharyngeal cancer. Alternatively, some team will find the appropriate targeted drug or immunotherapy, making the question of chemoradiation versus radiation moot. No specific funding was disclosed. Clifton D. Fuller has received grants from The University of Texas MD Anderson Cancer Center (project funding); GE Healthcare; Elekta AB; National Science Foundation, Division of Mathematical Sciences Quantitative Approaches to Biomedical Big Data (QuBBD) Program (NSF 1557559); and National Institutes of Health/National Institute for Craniofacial and Dental Research (1R56DE025248-01) as well as fellowship support from the National Cancer Institute Paul Calabresi Clinical Oncology Award (K12 CA088084), Speakers' Bureau and lecture fees from The University of Texas Health Science Center at San Antonio, publication royalties from Demos Medical Publishing, and accommodations/travel reimbursement for a Visiting Professor lecture from Oregon Health and Science University for work performed outside of the current study. David I. Rosenthal has received personal fees from Merck for work performed outside of the current study. Steven J. Frank has received a grant from and acted as a paid consultant for Varian; has received grants from the National Institutes of Health/National Cancer Institute and P01 and U19 grants; is employed by The University of Texas MD Anderson Cancer Center; has received honoraria from IBA and Hitachi; has Board membership for ABR, ABS, and C4 Imaging; and has had a patent issued for magnetic resonance imaging markers. Adam S. Garden, MD Department of Radiation Oncology The University of Texas MD Anderson Cancer Center Houston, Texas Clifton D. Fuller, MD, PhD Department of Radiation Oncology The University of Texas MD Anderson Cancer Center Houston, Texas David I. Rosenthal, MD Department of Radiation Oncology The University of Texas MD Anderson Cancer Center Houston, Texas William N. William Jr, MD Department of Thoracic Head and Neck Medical Oncology The University of Texas MD Anderson Cancer Center Houston, Texas Gary B. Gunn, MD Department of Radiation Oncology The University of Texas MD Anderson Cancer Center Houston, Texas Beth M. Beadle, MD, PhD Department of Radiation Oncology The University of Texas MD Anderson Cancer Center Houston, Texas Faye M. Johnson, MD Department of Thoracic Head and Neck Medical Oncology The University of Texas MD Anderson Cancer Center Houston, Texas William H. Morrison, MD Department of Radiation Oncology The University of Texas MD Anderson Cancer Center Houston, Texas Jack Phan, MD, PhD Department of Radiation Oncology The University of Texas MD Anderson Cancer Center Houston, Texas Steven J. Frank, MD Department of Radiation Oncology The University of Texas MD Anderson Cancer Center Houston, Texas Merrill S. Kies, MD Department of Thoracic Head and Neck Medical Oncology The University of Texas MD Anderson Cancer Center Houston, Texas Erich M. Sturgis, MD, MPH Department of Head and Neck Surgery; Department of Epidemiology The University of Texas MD Anderson Cancer Center Houston, Texas
Source: Cancer
DOI: 10.1002/cncr.30193
Language: English
Topic: Head and Neck Cancer Studies
Subfield: Otorhinolaryngology
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W3147343304