ORIGINAL ARTICLE IMPACT OF AGE AT DIAGNOSIS ON PROGNOSIS AND TREATMENT IN LARYNGEAL CANCER

Transcription

1 ORIGINAL ARTICLE IMPACT OF AGE AT DIAGNOSIS ON PROGNOSIS AND TREATMENT IN LARYNGEAL CANCER Johan A. Reizenstein, MD, 1 Stefan N. Bergström, MD, PhD, 4 Lars Holmberg, MD, PhD, 6 Arne Linder, MD, PhD, 3 Simon Ekman, MD, PhD, 2 Erik Blomquist, MD, PhD, 2 Britta Lödén, MD, 5 Marit Holmqvist, BSc, 7 Karin Hellström, RN, 7 Christer O. Nilsson, MD, 8 Daniel Brattström, MD, PhD, 2 Michael Bergqvist, MD, PhD 2 1 Department of Oncology, Örebro University Hospital, Örebro, Sweden. johan.reizenstein@orebroll.se 2 Department of Oncology, Uppsala University Hospital, Uppsala, Sweden 3 Department of Otorhinolaryngology and Head and Neck Surgery, Uppsala University Hospital, Uppsala, Sweden 4 Department of Oncology, Central Hospital, Gävle, Sweden 5 Department of Oncology, Central Hospital, Karlstad, Sweden 6 King s College London, School of Medicine, Division of Cancer Studies, London, UK and Regional Oncologic Centre, Uppsala, Sweden 7 Regional Oncologic Centre, Uppsala, Sweden 8 Department of Otorhinolaryngology and Head and Neck Surgery, Örebro University Hospital, Örebro, Sweden Accepted 3 September 2009 Published online 1 December 2009 in Wiley InterScience ( DOI: /hed Abstract: Background. The aims of this study were to analyze how age affects treatment and treatment outcome, and to determine whether tumor characteristics differ between different age groups with laryngeal cancer. Methods. Patients with laryngeal cancer during in the Uppsala Örebro region in Sweden were retrospectively studied. Results. There were no significant differences in the 945 cases between age groups concerning major patient and tumor characteristics, such as male/female ratio, distribution of Correspondence to: J. A. Reizenstein Contract grant sponsor: Cancer Foundation at Gävle Hospital; contract grant sponsor: Research Fund at the Department of Oncology, Uppsala University Hospital; contract grant sponsor: Larynx Foundation, Sweden. VC 2009 Wiley Periodicals, Inc. glottic/supraglottic tumors, stage, or site of recurrence. Overall survival (OS) and disease-specific survival (DSS) were worse among the oldest, although a significant proportion was cured. Relapse risk was lower among the oldest (12%) compared with the youngest (23%). The risk of never becoming tumorfree was 25% among the oldest and 7% in the youngest. Among the most elderly, only 1 late recurrence occurred. Conclusion. Elderly patients with laryngeal carcinoma cope well with treatment. Undertreatment may determine outcome more than age. The oldest group should be followed for a minimum of 2 years. VC 2009 Wiley Periodicals, Inc. Head Neck 32: , 2010 Keywords: laryngeal; cancer; age; prognosis; radiotherapy The question of whether age should be 1 of the determinants of treatment strategy is increasingly relevant in cancer care. In many countries, 1062 Laryngeal Cancer and Age HEAD & NECK DOI /hed August 2010

2 the rapidly growing elderly population entails an increased number of cancer patients, who also will be healthier than before. The life expectancy of a 70-year-old person is 15 years or more in many industrialized countries, and there will also be a significant growth in the number of very old (ie, people aged 85 years). There are limited and conflicting data as to the survival of elderly patients diagnosed with laryngeal carcinoma. Elderly patients are underrepresented in clinical trials, 1 leaving the clinician with poor guidance concerning the treatment for this patient category. Thus, specialists in otolaryngology and oncology face a dilemma when counseling elderly patients as to which is the optimal management. The objective of this study was to analyze how age has influenced treatment and treatment outcome, to determine whether tumor characteristics differ between age groups, and to collect data that could provide guidelines for the follow-up of elderly patients with laryngeal carcinoma. The source of information was a retrospective review of individual patient s charts in a large consecutive population-based series of 945 cases of laryngeal cancer, diagnosed during in 1 Swedish health care region. MATERIALS AND METHODS Sweden is divided into 6 health care regions, and the Uppsala Örebro region, which was used as the base population for this study, was the largest of these during the time period studied, with approximately 2 million inhabitants. In the region, there are 5 medical centers treating patients with laryngeal cancer. In Sweden it is mandatory by law to report all cases of cancer to the national cancer registry (NCR). We identified patients with invasive laryngeal cancer in the Uppsala Örebro region through the NCR. The medical records and radiotherapy charts for patients diagnosed during the years were retrospectively reviewed. Since 1998 there has been a prospective registration in a clinical database for all cases of laryngeal cancer at the Regional Oncologic Centre in the Uppsala Örebro region, and these patients were also included in the study and relevant data about these patients were transferred from the clinical database. The Swedish Cancer Registry registered 1004 patients with laryngeal cancer in our region during and clinical data were found for 961 patients, thus encompassing 96% of all cases. In 945 of these patients, follow-up data were possible to obtain, and these patients constitute the population of the analysis. All patients were clinically classified according to the TNM system (the Union Internationale Contre le Cancer [UICC] classification valid at the time of diagnosis). Characteristics age at the time of diagnosis, sex, histological diagnosis, tumor site, and treatment data (surgery, radiotherapy, and chemotherapy) were registered. The time of diagnosis was divided into 3 periods: , a total of 278 patients (30%) ; 336 patients (35%); and , 331 patients (34%). All patients were further divided into 4 groups based on age at the time of diagnosis: patients <50 years, 60 (6%); patients years, 473 (50%); patients years, 298 (32%); and patients 80 years, 114 (12%). Patients were initially followed approximately 1 month after radiation treatment and then generally every third month up to 2 years, and after that twice yearly up to 5 years or to relapse. Locoregional or distant relapses or deaths were recorded and considered events of interest. The cause of death was provided from the National Cause of Death Register held at the Swedish National Board of Health and Welfare. Statistics. Patients characteristics at diagnosis are presented with descriptive statistics. Overall survival (OS), disease-specific survival (DSS), and recurrence-free survival (RFS, time to recurrence) were analyzed with Kaplan Meier technique. The follow-up time was calculated from the date of diagnosis to event or last follow-up until December 31, In the analyses of DSS, the death from a cause other than laryngeal cancer was treated as a censored observation from time of death. Age was calculated as age at diagnosis. The endpoint DSS was analyzed in Cox regression models. Univariate and multivariate analyses were performed. Both univariate and multivariate models were adjusted by sex and age at diagnosis. Results are presented as hazard ratios with 95% confidence intervals (95% CI). RESULTS A total of 945 patients with laryngeal carcinoma were included, and of these 101 were women. Laryngeal Cancer and Age HEAD & NECK DOI /hed August

3 Table 1. Patients, tumor, and treatment characteristics by age group. Characteristic 50 y y y 80 y Sex Male 51 (85) 419 (89) 272 (91) 102 (89) Female 9 (15) 54 (11) 26 (9) 12 (11) Location Supraglottic 10 (17) 107 (23) 49 (17) 20 (18) Glottic 47 (78) 352 (75) 238 (80) 88 (78) Subglottic 3 (5) 10 (2) 9 (3) 5 (4) Classification Glottic T1 19 (32) 182 (38) 121 (41) 42 (37) T3/T4 19 (32) 142 (30) 73 (25) 41 (36) N0 51 (85) 417 (88) 268 (90) 101 (89) Nþ 9 (15) 56 (12) 30 (10) 13 (11) Treatment Curative RT 46 (77) 355 (75) 232 (78) 80 (70) (sole odality) Laryngectomy 5 (8) 65 (14) 19 (6) 4 (3) No treatment 0 (0) 9 (2) 14 (5) 13 (11) Abbreviation: RT, radiation therapy. No. (%) by age The median age and range for the decided time periods were: , median age 66 years (range, 35 89); , median age 68 years (range, 38 94); and , median age 69 years (range, 29 91). The differences in median age between the investigated time periods were not statistically significant (p ¼.084). In Table 1, further treatment characteristics and patients demographics are shown for the 4 age groups. There were only minor differences between the age groups in distribution of sex, tumor site, classification, and radiotherapy given with curative intent. Laryngectomy was performed less often in patients 80 years, and these patients tumors are more often not treated at all. In all, 20 patients received chemotherapy as part of their primary treatment. Numbers of relapses by age group are presented in Table 2. For example, in the youngest patient group there were 23% relapses, whereas in the elderly patient group 80 years there were 12%. However, patients never rendered tumor-free were among the elderly 25%, whereas in the youngest patient population the corresponding figure was 7%. Kaplan Meier estimates of OS are shown in Figure 1, death attributed to laryngeal carcinoma based on the Swedish National Board of Health and Welfare in Figure 2, and RFS in Figure 3. As expected, the OS rate (see Figure 1) was highly dependent on age, with a continuing risk of death throughout the period. The tumorspecific survival (see Figure 2) was considerably worse for the oldest patients, whereas the estimates with 95% CIs for the other age groups were similar. The 5-year DSS estimates for the different age groups were: <50 years, 0.83 ( ); years, 0.81 ( ); years, 0.76 ( ); and in patients 80 years, 0.61 ( ). However, in the oldest age group the RFS curve flattened out after 2 years of follow-up and RFS (Figure 3) was not worse for the elderly. Late recurrences, defined as >2 years after time of diagnosis, were scarce in the oldest population. The distribution of late recurrence in the different age groups were as follows: <50 years, 21.4% (3/14); years, 23.1% (24/104); years, 36.5% (19/52); and in patients 80 years, 7.1% (1/14). The results from the Cox regression analyses of DSS are presented in Table 3. The negative prognostic parameters in the multivariate model were age 80 years, higher T classification, and node-positivity. Tumor site, sex, and time period of diagnosis were not statistically significant in the multivariate model. DISCUSSION To our knowledge, the current report is the largest clinical study describing treatment allocated to and outcome of treatment of laryngeal cancer in elderly patients 80 years. Elderly patients with advanced tumors in the cohort studied were less often treated with radical surgery, and Table 2. Relapse pattern by age group. Age group No. Local relapse, % Regional relapse, % All relapse,* % Never-free-from-tumor, % <50 y y y y Total *Includes local, regional, distant, and unknown Laryngeal Cancer and Age HEAD & NECK DOI /hed August 2010

4 FIGURE 1. Kaplan Meier estimates of overall survival by age group. were more often offered only palliative care. The RFS rate for the elderly was higher than that for younger patients during the first 2 years of follow-up, but after that it flattened out and the risk of late recurrences for elderly patients with laryngeal carcinoma was in fact lower than that found for the younger patients. The present study is unique in terms of both the size of the cohort and its population-based character, as well as in the review of individual charts and a follow-up period for the majority of patients of >10 years. We aimed to include all the patients diagnosed with laryngeal carcinoma during in a well-defined geographical area to achieve a high degree of clinical applicability. During the study, data could be retrieved for 96% of the patients reported to the NCR during the period, untreated as well as treated. Reporting newly diagnosed cancer is mandatory in Sweden, and several studies have shown that Swedish cancer register data are of a high reliability; however, in a recent study an underreporting of 3.7% was found. 2 The low fraction of patients unavailable for analysis excludes a referral bias and vouches that the results represents a broad clinical practice. However, there are some limitations with the present study setting. The histopathological specimens were not reviewed for a second opinion. However, misclassification of tumors in this anatomical region is uncommon. Further, the TNM classification was based on the available imaging techniques during the different time intervals thus, there may be a stage migration with time and more sensitive staging techniques. The patients were not treated or followed in accord with a standardized protocol, but throughout the study period, the treatment and management were similar among the 5 centers. The majority of patients did not undergo autopsy, especially so among the elderly patients population, and thus the cause of death might FIGURE 2. Kaplan Meier estimates of disease-specific survival by age group. FIGURE 3. Kaplan Meier estimates of recurrence-free survival by age group. Laryngeal Cancer and Age HEAD & NECK DOI /hed August

5 Table 3. Risk ratios and 95% CIs estimated with Cox proportional hazards model for DSS (n ¼ 945). Model 1: Univariate Model 2: Multivariate Characteristic Risk ratio 95% CI Risk ratio 95% CI Sex Female 1.0 ( ) 1.4 ( ) Male 1.0 ref 1.0 ref Age at diagnosis, y < ( ) 0.5 ( ) ( ) 0.6 ( ) ref 1.0 ref ( ) 2.3 ( ) Inclusion period ref 1.0 ref ( ) 0.8 ( ) ( ) 0.9 ( ) Tumor origin Supraglottic 2.6 ( ) 1.0 ( ) Nonsupraglottic 1.0 ref 1.0 ref T classification T1 1.0 ref 1.0 ref T2 3.0 ( ) 3.1 ( ) T3 7.6 ( ) 6.8 ( ) T ( ) 8.1 ( ) N classification N0 1.0 ref 1.0 ref Nþ 4.6 ( ) 2.2 ( ) Radiation RT 60 Gy 2.2 ( ) 1.1 ( ) RT > 60 Gy 1.0 ref 1.0 ref No RT 1.4 ( ) 3.0 ( ) RT palliative 5.5 ( ) 3.2 ( ) Abbreviations: CI, confidence interval; DSS, disease-specific survival; RT, radiation therapy; Gy, gray. Note: Both models (univariate and multivariate) are adjusted for sex and age. The Cox model includes the following parameters: sex, age at diagnosis, inclusion period, tumor site, T classification, N classification, and radiation. Event ¼ disease-specific death ¼ 203 from 1978 to not be thoroughly investigated, which could lead to difficulties in estimating DSS data. We did not find a sharp increase with calendar time in the proportion of patients aged 80 years at diagnosis. There was only a modest trend for the median age of the patients to increase over time. Our data are not in accord with data from Jaseviciene et al, 3 who assessed the incidence of laryngeal cancer in Lithuania in the years , and showed that the mean age of male and female laryngeal cancer patients increased during the study period. Two recent meta-analyses on head and neck cancer, on radiotherapy 4 and on chemotherapy, 5 have both concluded that intensification of treatment may be detrimental to patients >70 years. However, many patients in the meta-analyses were included in clinical studies several years ago, and in recent years, radiotherapy, chemotherapy, and supportive care have all developed significantly. Allal et al 6 found that elderly patients (70 years) coped well with the treatment, in which the acute and late toxicities were similar to those observed in younger patients. Similar results were demonstrated by Pignon et al 7 in elderly patients with head and neck cancers treated with radiotherapy. Our study cannot directly address the cost benefit of radical treatment in the elderly, but we note that the worse DSS rates among the oldest may to a large extent be explained by a low treatment activity, since they were not disadvantaged by a worse stage distribution compared with that of the younger patients. Furthermore, the DSS data indicate that a substantial proportion of elderly can be cured and, if they survive over 2 years, local recurrences will be unlikely to impair quality of life. A clinical study by Colasanto et al 8 demonstrated that age did not significantly correlate with local relapse rates in patients with T1 and T2 laryngeal carcinoma; Sakata et al, 9 studying 1066 Laryngeal Cancer and Age HEAD & NECK DOI /hed August 2010

6 T1 and T2 glottic tumors, found that age did not significantly affect relapse-free survival; and for patients with head and neck squamous cell carcinoma, this is further shown by Italiano et al. 10 Contradictory to these results, in an univariate analysis, Lassaletta et al 11 found age to be a prognostic factor for relapse-free survival, as well as for OS and cause-specific survival. In the present study, there was a shift, so that failure with advancing age increasingly took the form of never-free-from-tumor rather than relapse. This may reflect a tendency to withhold therapy from the older patients. Several studies have previously demonstrated age to be a prognostic factor in laryngeal carcinoma, and in the present study age >80 years was associated with an increased risk of death ascribed to laryngeal cancer. In a study by Spector et al 12 on stage IV glottic carcinoma, patients younger than age 56 years had better survival (DSS) than patients 56 years of age (p ¼.002). Younger patients had less advanced T classification and more advanced neck disease compared with that of older patients (p ¼.004). Another study confirms age as a prognostic factor in stage IV disease. 13 The same tendency was observed in a study by Sessions et al 14 on supraglottic laryngeal carcinoma, in which improved DSS and cumulative DSS rates were associated with patients under the age of 65 years (p ¼.0001). In an American study by Nguyen-Tan et al, 15 age was found to be 1 of the significant prognostic factors for OS of advanced T3 T4 laryngeal carcinoma. In T3N0-1 glottic and transglottic carcinoma, Kowalski and colleagues 16 found age to be an important prognostic factor in patients who underwent surgery. Regardless of stage and region of origin, Johansen et al 17 demonstrated a better survival rate for younger patients (18 65 years). However, Teppo et al, 18 studying laryngeal cancer in Finland, found that the 5-year relative survival rate improved in both Northern Finland and the whole country, most noticeably among males and the elderly (65 years). In conclusion, this study indicates that age per se was used in the study population as a determinant for treatment strategy. The results of this study also indicate that elderly patients can be cured from laryngeal carcinoma, and suggest that they may benefit from conventional treatment. Late recurrence among the most elderly was a rare event. Based on this, we suggest that shorter and individualized follow-up periods for patients aged 80 years may be appropriate. Our findings indicate that patients aged 80 years should be followed for a minimum of 2 years, whereas younger patients should be followed for a minimum of 5 years. There were no significant differences between the age groups concerning major tumor characteristics, such as male/female ratio, the distribution of glottic/ supraglottic tumors, stage distribution, or the site of recurrence. Acknowledgments. The authors thank Michael Sihver and Bo Wilhelmsson for helpful assistance. REFERENCES 1. Horiot JC. Radiation therapy and the geriatric oncology patient. J Clin Oncol 2007;25: Barlow L, Westergren K, Holmberg L, Talback M. The completeness of the Swedish Cancer Register: a sample survey for year Acta Oncol 2009;48: Jaseviciene L, Gurevicius R, Obelenis V, Cicenas S, Juozulynas A. Trends in laryngeal cancer incidence in Lithuania: a future perspective. Int J Occup Med Environ Health 2004;17: Bourhis J, Overgaard J, Audry H, et al. Hyperfractionated or accelerated radiotherapy in head and neck cancer: a meta-analysis. Lancet 2006;368: Pignon JP, le Maitre A, Bourhis J. Meta-analyses of chemotherapy in head and neck cancer (MACH-NC): an update. Int J Radiat Oncol Biol Phys 2007;69(Suppl 2):S112 S Allal AS, Maire D, Becker M, Dulguerov P. Feasibility and early results of accelerated radiotherapy for head and neck carcinoma in the elderly. Cancer 2000;88: Pignon THJ, Van den Bogaert W, Van Glabbeke M, Scalliet P. No age limit for radical radiotherapy in head and neck tumours. Eur J Cancer 1996;12: Colasanto JM, Haffty BG, Wilson LD. Evaluation of local recurrence and second malignancy in patients with T1 and T2 squamous cell carcinoma of the larynx. Cancer J 2004;10: Sakata KAY, Karasawa K, Hasezawa K, et al. Radiation therapy in early glottic carcinoma: uni- and multivariate analysis of prognostic factors affecting local control. Int J Radiat Oncol Biol Phys 1994;30: Italiano A, Ortholan C, Dassonville O, et al. Head and neck squamous cell carcinoma in patients aged 80 years: patterns of care and survival. Cancer 2008;113: Lassaletta L, García-Pallares M, Morera E, Bernaldez R, Gavilan J. T3 glottic cancer: oncologic results and prognostic factors. Otolaryngol Head Neck Surg 2001;124: Spector GJ, Sessions DG, Lenox J, Newland D, Simpson J, Haughey BH. Management of stage IV glottic carcinoma: therapeutic outcomes. Laryngoscope 2004;8: Ampil FL, Nathan CAO, Caldito G, Lian TF, Aarstad RF, Krishnamsetty RM. Total laryngectomy and postoperative radiotherapy for T4 laryngeal cancer: a 14-year review. Am J Otolaryngol 2004;25: Laryngeal Cancer and Age HEAD & NECK DOI /hed August

7 14. Sessions DG, Lenox J, Spector GJ. Supraglottic laryngeal cancer: analysis of treatment results. Laryngoscope 2005;8: Nguyen-Tan PF, Le Q-T, Quivey J-M, et al. Treatment results and prognostic factors of advanced T3 4 laryngeal carcinoma: the University of California, San Francisco (UCSF) and Stanford University Hospital (SUH) experience. Int J Radiat Oncol Biol Physics 2001;50: Kowalski LP, Batista MBP, Santos CR, Scopel A, Salvajolli JV, Torloni H. Prognostic factors in T3, N0 1 glottic and transglottic carcinoma. A multifactorial study of 221 cases treated by surgery or radiotherapy. Arch Otolaryngol Head Neck Surg 1996;122: Johansen LV GC, Overgaard J. Laryngeal carcinoma multivariate analysis of prognostic factors in 1252 consecutive patients treated with primary radiotherapy. Acta Oncol 2003;42: Teppo HKP, Sipilä S, Jokinen K, et al. Decreasing incidence and improved survival of laryngeal cancer in Finland. Acta Oncol 2001;7: Laryngeal Cancer and Age HEAD & NECK DOI /hed August 2010