|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 19 (October 1), 2004: pp. 3981-3988 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.02.133 Matched-Pair Analysis of Survival of Never Smokers and Ever Smokers With Squamous Cell Carcinoma of the Head and NeckFrom The University of Texas M.D. Anderson Cancer Center, Houston, TX Address reprint requests to Erich M. Sturgis, MD, MPH, Department of Head and Neck Surgery, Unit 441, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030-4009; e-mail: esturgis{at}mdanderson.org
PURPOSE: To compare survival rates between patients with squamous cell carcinoma of the head and neck (SCCHN) without a history of smoking (never smokers) and those with a current or previous history of smoking (ever smokers). PATIENTS AND METHODS: Fifty never smokers with newly diagnosed SCCHN were matched to 50 ever smokers according to sex, age, tumor site, overall stage, nodal stage, and treatment. Survival analysis was performed using Kaplan-Meier estimates. Matched-pair survival was compared using the Cox proportional hazards model. RESULTS: The never smokers had a greater overall survival (P = .020), disease-specific survival (P = .022), and recurrence-free survival (P = .016). Furthermore, matched-pair analysis demonstrated smoking was associated with a significant increase in risk of overall death (relative risk [RR] = 3.50; 95% CI, 1.14 to 10.77; P = .029), risk of death owing to disease (RR = 3.98; 95% CI, 1.11 to 14.33; P = .034), and risk of disease recurrence (RR = 3.29; 95% CI, 1.18 to 9.14; P = .023). Smoking was associated with three-fold increases in risk for overall death, death owing to disease, and recurrence after adjustment for cancer-associated symptom severity and alcohol use, but the 95% CI for these adjusted risk estimates each included the null. CONCLUSION: Survival differed significantly between never smokers and ever smokers with SCCHN. These results are not substantively explained by differences in cancer-associated symptoms or alcohol use, but the CIs are wide and some imprecision remains. Regardless, possible fundamental differences in SCCHN between ever smokers and never smokers may exist, and further molecular characterization of these tumors is needed to determine whether biologic differences needing targeted therapies exist.
In the United States, more than 38,000 new cases of squamous cell carcinoma of the head and neck (SCCHN) will be diagnosed in 2004, and the overwhelming majority will be smoking-related.1 The relative risk of developing SCCHN is three to 12 times higher for smokers than for nonsmokers, and risk increases with increasing duration and dose of daily tobacco use.2-4 However, 4% to 10% of patients with SCCHN do not smoke, and they differ demographically from those patients who smoke (typically being younger and more commonly female).5-8 It has been postulated that nonsmokers who develop SCCHN have a genetic predisposition to cancer and thus are more susceptible to the development of SCCHN, even without tobacco exposure.9,10 Nonsmokers with SCCHN do have unique molecular changes in their tumors, but it is not known whether these changes affect tumor biology.9,11-13 Nonsmokers often present with earlier-stage disease, making direct comparison of survival between nonsmokers and smokers difficult.7,14-16 Furthermore, nonsmokers have fewer comorbidities than do smokers, which may improve survival. Although smoking is known to be a major risk factor for the development of SCCHN, the effect of smoking on prognosis is unclear.2 The unique demographics of nonsmokers with SCCHN further confound survival analysis. Matched-pair analysis allows for removal of many confounding factors and more accurate comparison of survival rates. We performed the first matched-pair analysis of nonsmokers and smokers with SCCHN and examined long-term survival to determine whether smoking has any effect on prognosis.
Patient Population More than 500 patients with newly diagnosed, previously untreated SCCHN were entered onto a prospective epidemiologic study at The University of Texas M.D. Anderson Cancer Center between 1995 and 2001; the study was approved by the institutional review board. Only incident cases of pathologically confirmed SCCHN were included, whereas those with salivary gland, nasopharyngeal, or lip carcinoma or those initially treated elsewhere were not recruited. Eighty-three patients were nonsmokers and were available for matching. We used this prospective epidemiologic database to establish our matched pair analysis, because all participants completed a questionnaire (as described later in this section) regarding tobacco use, allowing us to accurately classify our patients as never or ever smokers. After providing informed consent, all patients participating in this study prospectively completed at presentation a self-administered epidemiologic questionnaire that included demographic and exposure information. Former smokers were defined as smokers who had quit smoking at least 1 year before presentation, and former smokers were grouped with current smokers as ever smokers. Never smokers were defined as those who had smoked fewer than 100 cigarettes in their lifetime. Pharmacologic testing of current tobacco use using serum cotinine levels has previously verified high concordance with questionnaire data in this population. Drinkers were defined as those who had at least one alcoholic drink per week for at least 1 year, whereas former drinkers were defined as those who had quit dinking alcoholic beverages at least 1 year before presentation. Medical records were reviewed for primary site, clinical stage, treatment, histology, outcome measures, and medical and symptom comorbidities. Symptom comorbidities were classified according to the Cancer Associated Symptom Index, which scores the number of symptoms related to head and neck cancer prognosis.17 These include weight loss, dysphagia, neck mass, and otalgia. An index of these symptoms is scored as none (0 of four symptoms), mild (one of four symptoms), moderate (two of four symptoms), and severe (three of four or four of four symptoms). Medical comorbidities were classified according to the Kaplan Feinstein Comorbidity Factor, which reflects the presence of related comorbidities, from none to mild, moderate, or severe disease.18 Comorbid conditions include the following categories: cardiac, endocrine, hypertension, renal, respiratory, cerebral/psychologic, hepatic, locomotor, gastrointestinal, and peripheral vascular disease; alcoholism; other malignancies; and miscellaneous. These comorbidity scales have been demonstrated to be independent prognosticators in SCCHN.19 From this cohort, each never smoker was matched with one ever smoker. Matching variables were age (+ 10 years), sex, site of primary tumor (oral cavity, oropharynx, hypopharynx, or larynx), disease stage (stage I/II v stage III/IV), pathologic nodal status (positive v negative), and treatment received (surgery; radiation therapy; surgery and radiation therapy; chemotherapy and radiation therapy; or chemotherapy, radiation therapy, and surgery). All patients were treated for curative intent and were free of disease at the end of treatment. Fifty pairs of never smokers and ever smokers were matched.
Statistical Methods
Demographics The matched characteristics of the two groups are listed in Table 1. The patients were matched by age (+ 10 years), sex, disease site, overall stage (I/II v III/IV), nodal status, and treatment. As expected, there were no significant differences between the two groups with respect to the matching variables. The patients ranged in age from to 31 to 83 years, with a mean of 54.3 years (median, 51.0 years) for the never smokers and 55.2 years (median, 52.1 year) for the ever smokers (P = .70). The patient unmatched characteristics of the two groups are listed in Table 2. The makeup of the groups was similar for T stage, N stage, ethnic group, and Kaplan-Feinstein Comorbidity score but significantly different for alcohol status (P = .04) and Cancer Symptom Index (P = .04). Alcohol status and Cancer Symptom Index were therefore included in the multivariate analysis.
Among the 13 never smokers with oral cavity cancer, 10 cases were of the oral tongue, one case was of the floor of mouth, one case was of the gingiva, and one case was of the buccal mucosa. Among the 13 ever smokers with oral cavity cancer, three cases were of the oral tongue, four cases were of the floor of mouth, five cases were of the gingiva, and one case was of the hard palate. Among the 31 never smokers with oropharyngeal cancer, 15 cases were of the tonsil, 14 were of the base of tongue, one was of the pharyngeal wall, and one was of the soft palate. Among the 31 ever smokers with oropharyngeal cancer, 18 cases were of the tonsil, 12 were of the base of tongue, and one was of the pharyngeal wall. Both patients with hypopharyngeal cancer had primaries of the pyriform sinus. Of the five never-smoker patients with laryngeal cancer, four cases were glottic and one was supraglottic, whereas of the ever-smoker patients, three were glottic and two were supraglottic. As compared with the 50 never smokers included in the study, the unmatched never smokers (n = 33) were statistically more likely to have more advanced tumor stage (P = .018), to have nodal metastases (P = .041), and to have tumors in the oral cavity (P = .018). As expected given the more advanced disease stage, the unmatched group of never smokers had poorer 2-year and 5-year survival rates compared with the never smokers in the study (86.6% v 95.1% and 64.6% v 83.2%, respectively). However, the study was designed to avoid selection biases by comparing groups tightly matched on the major prognostic, demographic, and treatment variables.
Of the 50 ever smokers, 29 were former smokers and 21 were smoking at the time of presentation. Of the former smokers, eight had quit
Survival Analysis Fifteen of the 100 patients died (three never smokers and 12 ever smokers). Of these 15 patients, 14 died from disease (three never smokers and 11 ever smokers). Figure 1 shows the Kaplan-Meier overall survival curve for the never smokers versus the ever smokers. There was a significant difference (P = .020) in mean overall survival among the never smokers and ever smokers. Figure 2 shows the Kaplan-Meier disease-specific survival curve for the ever smokers versus the never smokers. The mean survival until death owing to disease significantly differed (P = .022) between the never smokers and ever smokers. Nineteen of the patients had a recurrence event; five were never smokers and 14 were ever smokers. Figure 3 shows the Kaplan-Meier recurrence-free survival curves for the never smokers and ever smokers. There was a significant difference (P = .017) in recurrence-free survival between the never smokers and ever smokers.
Matched-Pair Analysis In a matched-pair study, each pair is classified by the pattern of outcome events experienced within the pair. Pairs in which each patient experienced the same event are classified as concordant, and pairs in which one patient experienced an event and the other did not are said to be discordant. There were two concordant pairs in which both the ever smoker and never smoker died, 11 discordant pairs in which the ever smoker died and the never smoker did not, and two discordant pairs in which the never smoker died and the ever smoker did not. Smoking was associated with a significantly increased risk of overall death (RR = 3.50; 95% CI, 1.14 to 10.77; P = .029). Multivariate analysis was performed for those factors that were significantly different between the two groups (Cancer Associated Symptom Index and alcohol use). Although ever smokers had three times the risk of overall death compared with never smokers after multivariate adjustment, the 95% CIs were wide, reflecting some imprecision but showing that the data are compatible, with a range of effect sizes from roughly the null to a strong, increased risk of death (Table 3).
There was one concordant pair in which both the never smoker and ever smoker died from disease, 10 discordant pairs in which the ever smoker died from disease and the never smoker did not, and two discordant pairs in which the never smoker died from disease and the ever smoker did not. There was a statistically significant increase in risk of death owing to disease (relative risk [RR] = 3.98; 95% CI, 1.11 to 14.33; P = .034). Multivariate analysis was performed, adjusting for alcohol use and Cancer Associated Symptom Index, and the risk of death owing to disease was three times higher in the ever smokers. Although the 95% CIs were wide, reflecting some imprecision, they did include effect sizes from roughly no effect to a strong risk of death from disease (Table 3). There were two concordant pairs in which both the never smoker and the ever smoker had a disease recurrence, 12 discordant pairs in which the ever smoker had a disease recurrence and the never smoker did not, and three discordant pairs in which the never smoker had a disease recurrence and the ever smoker did not. There was a statistically significant increase in risk of recurrence (RR = 3.29; 95% CI, 1.18 to 9.14; P = .023), and this risk among ever smokers remained three times that of never smokers after further adjustment for alcohol use and Cancer Associated Symptom Index (Table 3). Although the adjusted 95% CI was similar in width, it included a range of risk estimates from the null to a strong risk of recurrence.
There are no previously reported matched-pair analyses examining the effect of smoking on survival in patients with SCCHN. We matched 50 never smokers to 50 ever smokers, using the matching variables of age, sex, site, disease stage, pathologic nodal status, and treatment. We found that ever smokers had lower overall survival, disease-specific survival, and recurrence-free survival than did matched never smokers of the same disease stage who underwent the same treatment. This finding was no longer statistically significant after adjustment for alcohol use and Cancer Associated Symptom Index. Although the 95% CIs were wide after adjustment, reflecting some imprecision, they did suggest a worse outcome for smokers and a range of effect sizes from approximately the null to a strong risk of death, of death owing to disease, and of recurrence. The association between smoking and the development of SCCHN has been demonstrated repeatedly, but the effect of smoking on survival has not been as well addressed.2 Koch et al13 studied 305 patients with SCCHN, 46 of whom were nonsmokers, and did not find a statistically significant difference in overall survival between nonsmokers and smokers. SCCHN specimens from nonsmokers have molecular changes that are different from those of smokers.11-13,21 p53 mutations are more common in smokers than in nonsmokers who do not drink alcohol.22 The published findings on the prognostic significance of p53 mutations are conflicting, and the conclusions of most of these studies are limited because they used immunohistochemical techniques, had small sample sizes, or both. However, one larger series showed that patients with p53 mutations have a worse prognosis than those without p53 mutations.23 Other studies have suggested that human papillomavirus (HPV) type 16 may play a major role in oropharyngeal cancer in nonsmokers, and several studies reported that patients with tumors positive for HPV have a better prognosis than do those with HPV-negative tumors.13,20,24-27 These findings suggest that SCCHN in smokers may be a different disease at the molecular level. It is therefore not unreasonable to find differences in their clinical behavior. These molecular changes may cause a more aggressive disease in smokers and account for their worse prognosis. Although our treatment paradigms have not substantively changed during the timeframe of this study and we have matched our two groups on treatment, it is possible that some differences in treatment remain that could contribute to the outcome differences. Because the two groups are matched on treatment, the same number of oral cavity cancer patients in each group were treated by surgery alone and by surgery with postoperative radiotherapy (the indications for which have not changed during this time period). Although we did not match for the inclusion of a neck dissection as part of their treatment, a similar number of oral cavity patients in each group had a neck dissection as part of their treatment (nine never smokers and eight ever smokers; P = .680). Again, because the two groups were matched on treatment, the same number of oropharyngeal cancer patients in each group were treated by radiotherapy alone and by radiotherapy with chemotherapy. Although we did not match for radiotherapy dose, the radiation fractionation schema, or sequence of chemotherapy, these treatment variables were used in a similar number of oropharyngeal patients in the two groups. For instance, the average dose of radiotherapy received was 71.0 ± 2.2 for the never smokers and 71.2 ± 1.8 Gy for the ever smokers (P = .654). Secondly, 23 of the never smokers received concomitant boost radiotherapy, as compared with 24 of the ever smokers (P = .767; the remaining oropharyngeal cancer patients received conventional fractionation). No patient received intensity-modulated radiotherapy. Six of the never smokers received concurrent chemotherapy, as compared with 10 of the ever smokers (P = .393). An additional six never smokers received induction chemotherapy, and two ever smokers received induction chemotherapy (P = .130). During this time period, a randomized placebo-controlled phase III trial using adjuvant antiepidermal growth factor receptor antibody (C225) was used for oropharyngeal cancer patients treated with radiotherapy. These patients were coded as receiving concurrent chemotherapy if they received C225 and radiation only if they received placebo. Four never smokers and four ever smokers received C225. One additional never smoker patient was miscoded as receiving placebo (ie, radiation only), when in fact the patient received radiation with C225. However, as this is untested adjuvant therapy to radiation for oropharyngeal cancer, we do not feel this has significantly biased our results. Additionally, we have found no significant differences in any of the above treatment specifics for the six pairs of hypopharyngeal/laryngeal cancer patients. Two highly selected phase II experimental treatment protocols using induction chemotherapy before surgery for oral tongue cancer in younger adults or in laryngeal cancer of intermediate stage were developed during this timeframe. Because of the limited numbers enrolled in these protocols and the tight matching requirements for this study, no patient enrolled in these protocols was included in this matched-pair analysis. Although some differences in treatment specifics do exist between the two groups, none of these are statistically significant, and we feel that it is unlikely they have had an appreciable impact on our findings. Medical and symptom comorbidities are known prognostic factors of SCCHN.17,19 We found no significant difference between the ever smokers and never smokers in the number of patients who had moderate to severe medical comorbidities (P = .338). This was unexpected, given the association of smoking with coronary artery disease, peripheral vascular disease, and chronic obstructive pulmonary disease. However, there was a statistically significant difference in Cancer Associated Symptom Index scores. That the elevated symptom index may indicate a more aggressive disease in ever smokers is supported by the poorer survival. Performance status, which may in part be related to one's smoking history, is also a predictor of outcome for patients with head and neck cancer, especially for those receiving multimodality therapy that includes chemotherapy.28,29 Our study included only patients with local or locoregional disease treated for cure, and they were treated primarily by surgeons and radiation oncologists (74%). Consequently, Eastern Cooperative Oncology Group performance status or Karnofsky performance status was not documented for every patient, precluding the inclusion of performance status in the analyses. However, it is likely that the Cancer Associated Symptom Index and Kaplan Feinstein Comorbidity Index are surrogate markers of performance status. The two groups had similar Kaplan Feinstein Comorbidity Indices, but the group of ever smokers did have a worse Cancer Associated Symptom Index, and consequently, this factor was included in the multivariate modeling. Again, this elevated index suggests a more aggressive disease in the ever smokers, which is supported by the poorer survival in the ever smokers. Heavy alcohol use is often associated with smoking but may also be linked to survival and performance status. We did find smokers to be more commonly users of alcohol, and therefore alcohol use as well as Cancer Associated Symptom Index was included in the multivariate models. It is possible that residual confounding owing to heavy alcohol use could exist, and other confounders may exist that were not controlled for. For instance, histologic factors such as extracapsular spread and perineural invasion could not be analyzed because we did not have this information on most patients (only one third underwent surgery). Regardless, patients' symptoms, tobacco/alcohol use, and comorbidities must be aggressively managed to optimize the patients' quality of life and survival. Several studies have been designed to show the effect of continued smoking after diagnosis on outcome. Browman et al30 studied patients undergoing radiotherapy for SCCHN and initially found that those who continued to smoke during therapy had decreased survival compared with those who stopped smoking. However, follow-up studies were unable to show any significant survival difference.31,32 Certainly, those who continue to smoke after diagnosis likely have a higher rate of second primary malignancy development, again underscoring the importance of smoking cessation efforts. We chose to compare never smokers with ever smokers, rather than compare nonsmokers who may have a history of smoking with current smokers, to clearly determine the effect of smoking (past or present) on survival. Although we did not see a difference in the 2-year survival rates between current and former smokers, the present study was not designed for this purpose. A matched-pair analysis comparing the survival of current and former smokers or the different levels of exposure is needed. It is also possible that selection biases affecting either the patients referred to our institution or the patients recruited to our epidemiologic study could exist. Although in the initial years of our epidemiologic study not all patients with incident SCCHN presenting to our institution were invited to participate, this was due to manpower limitations at that time rather than any selection. Furthermore, more than 95% of patients invited to participate have been enrolled. Finally, we have no evidence that a differential (prognostically) selection of patients to our matched analysis would exist between never smokers and ever smokers. To overcome limitations of small sample size, power, and selection biases, we chose a matched-pair analysis, matching for age, sex, site, disease stage, nodal status, and treatment. In this way, we avoided comparisons of never smokers and ever smokers with different demographic makeup or clinical characteristics and matched patients for many of the factors known to affect prognosis. The power of this study to detect a three-fold difference in risk was greater than 80%, and therefore a sample size of 50 was adequate. This power calculation does not include the additional statistical precision afforded by the extensive matching process. In conclusion, ever smokers diagnosed with SCCHN had shorter overall, disease-specific, and recurrence-free survival compared with never smokers of equivalent age, sex, site, disease stage, nodal status, and treatment at our institution. These results are not substantively explained by differences in cancer-associated symptoms or alcohol use, but the CIs are wide and some imprecision remains. Although our findings suggest that SCCHN in ever smokers and SCCHN in never smokers may have more fundamental differences, further studies are needed to clarify the molecular differences between SCCHN in ever smokers and never smokers and the effect of smoking on prognosis. Patients' comorbidities, tobacco/alcohol use, and symptoms should be aggressively managed to optimize their quality of life and survival.
The authors indicated no potential conflicts of interest.
We thank Maureen Goode, PhD, for scientific editing, Deanna Thomas for manuscript preparation, and Adam S. Garden, MD, for review of radiotherapy schema.
Presented as a poster presentation at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Jemal A, Tiwari RC, Murray T, et al: Cancer statistics, 2004. CA Cancer J Clin 54:8-29, 2004
2. Blot WJ, McLaughlin JK, Winn DM, et al: Smoking and drinking in relation to oral and pharyngeal cancer. Cancer Res 48:3282-3287, 1988 3. Diaz EM, Sturgis EM, Laramore GE, et al: Neoplasms of the head and neck, in Holland JF, Frei E (eds): Cancer Medicine (ed 6). Hamilton, Ontario, Canada, BC Decker Inc, 2003, pp 1325-1371 4. Lewin F, Norell SE, Johansson H, et al: Smoking tobacco, oral snuff, and alcohol in the etiology of squamous cell carcinoma of the head and neck: A population-based case-referent study in Sweden. Cancer 82:1367-1375, 1998[CrossRef][Medline]
5. Muscat JE, Richie JP Jr, Thompson S, et al: Gender differences in smoking and risk for oral cancer. Cancer Res 56:5192-5197, 1996 6. Bundgaard T, Bentzen SM, Wildt J, et al: Histopathologic, stereologic, epidemiologic, and clinical parameters in the prognostic evaluation of squamous cell carcinoma of the oral cavity. Head Neck 18:142-152, 1996[CrossRef][Medline] 7. Agudelo D, Quer M, Leon X, et al: Laryngeal carcinoma in patients without a history of tobacco and alcohol use. Head Neck 19:200-204, 1997[CrossRef][Medline] 8. Schantz SP, Byers RM, Goepfert H, et al: The implication of tobacco use in the young adult with head and neck cancer. Cancer 62:1374-1380, 1988[CrossRef][Medline] 9. Cheng L, Sturgis EM, Eicher SA, et al: Glutathione-S-transferase polymorphisms and risk of squamous-cell carcinoma of the head and neck. Int J Cancer 84:220-224, 1999[CrossRef][Medline] 10. Lingen M, Sturgis EM, Kies MS: Squamous cell carcinoma of the head and neck in nonsmokers: Clinical and biologic characteristics and implications for management. Curr Opin Oncol 13:176-182, 2001[CrossRef][Medline] 11. Sturgis EM, Wei Q: Genetic susceptibility: Molecular epidemiology of head and neck cancer. Curr Opin Oncol 14:310-317, 2002[CrossRef][Medline] 12. Koch WM, McQuone S: Clinical and molecular aspects of squamous cell carcinoma of the head and neck in the nonsmoker and nondrinker. Curr Opin Oncol 9:257-261, 1997[Medline] 13. Koch WM, Lango M, Sewell D, et al: Head and neck cancer in nonsmokers: A distinct clinical and molecular entity. Laryngoscope 109:1544-1551, 1999[CrossRef][Medline] 14. Wey PD, Lotz MJ, Triedman LJ: Oral cancer in women nonusers of tobacco and alcohol. Cancer 60:1644-1650, 1987[CrossRef][Medline] 15. Trigg DJ, Lait M, Wenig BL: Influence of tobacco and alcohol on the stage of laryngeal cancer at diagnosis. Laryngoscope 110(3 Pt 1):408-411, 2000 16. de Boer MF, Sanderson RJ, Damhuis RA, et al: The effects of alcohol and smoking upon the age, anatomic sites and stage in the development of cancer of the oral cavity and oropharynx in females in the southwest Netherlands. Eur Arch Otorhinolaryngol 254:177-179, 1997[CrossRef][Medline] 17. Pugliano FA, Piccirillo JF, Zequeira MR, et al: Symptoms as an index of biologic behavior in head and neck cancer. Otolaryngol Head Neck Surg 120:380-386, 1999[CrossRef][Medline] 18. Kaplan MH, Feinstein AR: The importance of classifying initial co-morbidity in evaluating the outcome of diabetes mellitus. J Chronic Dis 27:387-404, 1974[CrossRef][Medline] 19. Hall SF, Groome PA, Rothwell D: The impact of comorbidity on the survival of patients with squamous cell carcinoma of the head and neck. Head Neck 22:317-322, 2000[CrossRef][Medline] 20. Dupont WD, Lummer WD: Power and sample size calculations: A review and computer program. Control Clin Trials 11:116-128, 1990[CrossRef][Medline] 21. Sisk EA, Soltys SG, Zhu S, et al: Human papillomavirus and p53 mutational status as prognostic factors in head and neck carcinoma. Head Neck 24:841-849, 2002[CrossRef][Medline]
22. Brennan JA, Boyle JO, Koch WM, et al: Association between cigarette smoking and mutation of the p53 gene in squamous-cell carcinoma of the head and neck. N Engl J Med 332:712-717, 1995 23. Pukkila MJ, Kumpulainen EJ, Virtaniemi JA, et al: Nuclear and cytoplasmic p53 expression in pharyngeal squamous cell carcinoma: Prognostic implications. Head Neck 24:784-791, 2002[CrossRef][Medline]
24. Dahlstrom KR, Adler-Storthz K, Etzel CJ, et al: Human papillomavirus type 16 infection and squamous cell carcinoma of the head and neck in never smokers: A matched pair analysis. Clin Cancer Res 9:2620-2626, 2003
25. Fouret P, Monceaux G, Temam S, et al: Human papillomavirus in head and neck squamous cell carcinomas in nonsmokers. Arch Otolaryngol Head Neck Surg 123:513-516, 1997 26. Ritchie JM, Smith EM, Summersgill KF, et al: Human papillomavirus infection as a prognostic factor in carcinomas of the oral cavity and oropharynx. Int J Cancer 104:336-344, 2003[CrossRef][Medline] 27. Schwartz SR, Yueh B, McDougall JK, et al: Human papillomavirus infection and survival in oral squamous cell cancer: A population-based study. Otolaryngol Head Neck Surg 125:1-9, 2001[CrossRef][Medline] 28. Pivot X, Niyikiza C, Poissonnet G, et al: Clinical prognostic factors for patients with recurrent head and neck cancer: Implications for randomized trials. Oncology 61:197-204, 2001[CrossRef][Medline] 29. Final Report of the Head and Neck Contracts Program. Adjuvant chemotherapy for advanced head and neck squamous carcinoma. Cancer 60:301-311, 1987[CrossRef][Medline]
30. Browman GP, Wong G, Hodson I, et al: Influence of cigarette smoking on the efficacy of radiation therapy in head and neck cancer. N Engl J Med 328:159-163, 1993 31. Browman GP, Mohide EA, Willan A, et al: Association between smoking during radiotherapy and prognosis in head and neck cancer: A follow-up study. Head Neck 24:1031-1037, 2002[CrossRef][Medline] 32. Franchin G, Minatel E, Gobitti C, et al: Radiotherapy for patients with early-stage glottic carcinoma: Univariate and multivariate analyses in a group of consecutive, unselected patients. Cancer 98:765-772, 2003[CrossRef][Medline] Submitted February 20, 2004; accepted July 26, 2004.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 test and the two-tailed Fisher's exact test to detect any significant differences between never smokers and ever smokers. Those factors, which were significantly different, were used as variables in the Cox proportional hazards model. The relative risk was obtained for each type of event between never smokers and the ever smokers. The power of the study to detect the relative risk obtained was based on the model of Dupont and Lummer.
5 years before presentation, whereas 21 had quit more than 5 years before presentation. Six of the former smokers had smoked less than 10 pack-years, six had smoked 10 to 20 pack-years, and 16 had smoked for more than 20 pack-years. Of the current smokers, only one had a 10 to 20 pack-year history of smoking, whereas 19 had a history of smoking more than 20 pack-years. One current smoker and one former smoker were unwilling to give information as to the number of pack-years they had smoked. There is no standardized documentation of continued smoking available for the 21 current smokers, but a review of the medical records showed that 50% continued to smoke after treatment began. Although this analysis was not designed to determine whether there was a difference in survival between former and current smokers, the 2-year overall survival rates were similar (78.6% and 77.3%, respectively). Among the smokers providing pack-year information, a trend of worsening survival was suggested with greater pack-years of exposure; however, the study was not designed to compare the survival of these small exposure subgroups, which are unmatched for prognostic variables. 
