Prognostic factors in oral tonguecancer – does age, gender or timeperiod of treatment predictoutcome?

(1)

Örebro University School of Medicine Degree project, 30 ECTS June 2017

Prognostic factors in oral tongue

cancer – does age, gender or time

period of treatment predict

outcome?

Version 2

Author: Oscar Smedberg, bachelor of medicine

Supervisors: Fredrik Landström MD, PhD

(2)

Abstract

Introduction. Oral cavity cancer, including its sub-site oral tongue cancer (OTC) is a major contributor to the global cancer burden. It mainly afflicts the elderly population but ~5% is below 40 years old. The incidence is rising and some reports suggest that young patients with OTC are inclined to a lower survival compared to older patients. Aim. Compare overall survival (OS) based on patient age (≤40 or >40 years), gender, disease stage, and diagnosis year (<2000 or ≥2000).

Materials and methods. This was a retrospective observational study. Data for all cases of squamous cell carcinoma (SCC) of the oral tongue treated at Örebro

University Hospital during the years 1988-2014 was extracted from the Örebro Head and Neck Cancer Registry. In addition, medical records were reviewed for smoking status and histological grading. The Kaplan-Meier method was used for survival analyses. In all, 342 patients were included in this study.

Results. Median age at diagnosis was 64 years (range 20-91). 29 patients (8.5%) were aged ≤40 years. 5-year OS was 75.6% and 50.3% for the young and old group

respectively. Diagnosis after year 2000 was associated with a 36% improved survival. There was no difference in survival between genders.

Conclusion. Young adults (≤40 years) with SCC of the oral tongue had a significantly higher 5-year OS than old patients (>40 years). These results do not support the hypothesis that young patients with OTC have a lower survival than older patients.

Keywords: Oral tongue cancer, squamous cell carcinoma, young adults, survival comparison

(3)

Abbrevations

HPV – human papillomavirus M:F – male to female

OP – old patients (>40 years old) OS – overall survival

OTC – oral tongue cancer SCC – squamous cell carcinoma YP – young patients (≤40 years old)

(4)

1. Introduction

Oral cavity cancer varies in incidence in different regions of the world but is one of the most common cancers worldwide [1–3]. A sub-site, oral tongue cancer (OTC), represents 22% to 49% of all oral cavity cancers[3,4]. Oral tongue is defined as the front two thirds of the tongue and thus, does not include the base of tongue [5].

OTC, of which squamous cell carcinoma (SCC) represents >95% of cases [6], is mainly a disease of the elderly population with a median age of ~60 years at diagnosis [7–9]. Only ~5% of the patients are younger than 40 years [10,11].

The incidence of oral cavity cancer shows a varying male dominance across the world ranging from 1.4:1 for northern Africa, western Asia, and Oceania to 5.2:1 for central and eastern Europe [12]. A few smaller studies have reported a male to female (M:F) ratio of 1.8 in Australia [7], 2.9 in Germany [8], and 1.6 in the Nordic countries [10]. Some reports also suggest that male gender is associated with lower survival for oral cavity cancer in general [8] and OTC more specifically [13,14].

Sweden has among the lowest incidence rate of oral cancer in Europe[15], with 177 new cases of OTC reported in 2014 [16]. In two large epidemiological studies

Annertz et al. have shown an increase in the incidence of oral cancer, especially OTC, in the Nordic countries, particularly in young adults [6,10]. These results mirror the development seen in other European countries as well as parts of the United States [15]. The reason for the increasing incidence is mainly unknown. Smoking and alcohol consumption, the traditional main risk factors for tongue cancer, are declining in both Europe and the USA [6] and have shown to be of limited importance to the development of oral cavity cancer in young patients [17]. A hypothesis associating infection with oncogenic strains of human papillomavirus (HPV) to OTC have been tested but such a connection, although examined in several studies, have yet to be proved [4,9,18].

(6)

surgical resection alone is used. For late stages with suspected distant metastasis, palliative chemotherapy is sometimes indicated. Because OTC commonly

metastasises to regional lymph nodes, >40% for T2 lesions or larger, a neck

dissection is often performed. Despite this, locoregional recurrences account for 60-70% of deaths in OTC[1].

In spite of recent advances in diagnosis and cancer treatment, no major improvement in survival rates for OTC have been seen and overall 5-year survival for these cancers ranges from ~70% for stage I disease to ~15% for stage IV [1]. Hammarstedt et al. showed only a modest improvement on age-standardized relative survival for Swedish tongue cancer patients during the period 1960-2004 [19]. Results also suggest that SCC of the oral tongue has a worse prognosis than SCCs in other parts of the oral cavity [20].

Whether the outcome for younger patients (≤40 years) with OTC is worse than for older patients remains unresolved. Several studies have previously investigated this with conflicting results. Most studies [7,13,14,21–23] have reported a similar overall survival (OS) while others suggest a poorer outcome for young adults [24,25]. Three of the studies also reported a higher recurrence rate for the young group, even though it did not always translate to a lower OS [7,24,25]. Most of these studies were small and case-matched or had the study group composed of otherwise specifically selected cases with varying statistical significance, and none included a Nordic population. The aim of the present study was to compare the OS and recurrence rates for young patients (YP, ≤40 years) and old patients (OP, >40 years). In addition the OS with regards to gender, tumour stage and if the patient was diagnosed before or after the year 2000 was investigated.

2. Materials and methods

2.1 Study population

The Örebro Head and Neck Cancer Registry includes all patients treated for head and neck cancers since 1988 at Örebro University Hospital. From the registry, all reported

(7)

cases of OTC for the years 1988-2014 were included. The time period was chosen to allow for at least a 2-year follow-up. The registry provided information about age, gender, date when a case was reported to the registry (this was used as date of diagnosis), TNM-classification and tumour stage (I-IV), histological type, date of recurrence and date of death. Recurrence events were categorized as being local, regional, distant or both local and regional. Information about tumour differentiation grade and patient smoking status (defined as current smoker, ex-smoker or non-smoker) were collected from patient

records.

The registry contained 348 cases for the selected time period. Two patients were found to be registered twice with the same ICD-code and the same cancer type. For each of these patients, the latter case was excluded. Of the remaining 346 cases, 342 (98.8 %) were SCCs and these were selected for the statistical analysis. The four cases with other tumour types (three

adenocarcinomas, one malignant melanoma) were excluded for a more homogenous patient group and more

comparable results. These four cases all came from the OP group. The 342 cases of SCCs included 163 males (48.3%) and 179 females. Figure 1 shows a flowchart of the selection process.

2.2 Treatment strategy

All treatment decisions were made by a multidisciplinary cancer team and according to local guidelines. This study has not taken into account any treatment given since this tend to depend highly on disease stage and spread, although in some part also

348 cases of oral tongue cancer found

in the registry.

342 cases were selected for analysis. 2 cases found to

be registered twice and were

excluded. 4 cases were other cancer types than

squamous cell carcinoma. 3 adenocarcinomas, 1 malignant melanoma.These were excluded.

(8)

physical state of a patient, and aim of this study was not to compare different treatment strategies.

2.3 Statistical analysis

Results were tested for normality using the Shapiro-Wilk test. Patient group characteristics were compared using chi-square test and likelihood ratio. Survival curves were calculated with the Kaplan-Meier method and were compared for statistical significance using the log-rank test. IBM SPSS Statistics for Macintosh, version 22 (IBM Corp., Armonk, N.Y., USA) was used for statistical analyses. A P-value of 0.05 was chosen as the statistical significance level.

3. Results

3.1 Survival based on age groups

Overall median age at diagnosis was 64 years (range 20-91). 29 patients (8.5%) were aged ≤40 at diagnosis. Median age for the YP and OP was 33 years and 65 years respectively. Patient characteristics were similar between the groups in regard to gender, TNM stage and recurrence events but smoking status and tumour

differentiation grade differed significantly for the two age groups, the YP group having a higher rate of never smokers (48.3% vs. 35.8%) and more highly differentiated tumours with no case of low differentiation grade as opposed to 43 (13.7%) in the OP group. For patient characteristics see table 1.

Cumulative OS at 2 and 5 years was 82.8% and 75.6% respectively for the YP. For the OP the OS were 61.3% at 2 years and 50.3% at 5 years. The survival difference was statistically significant (p=0.012). See figure 2.

Overall recurrence rate was 31.0% for YP and 33.5% for OP. The difference in recurrence was not statistically significant (p=0.418). See table 2 for details.

(9)

Table 1. Characteristics of the study group stratified for age Characteristic ≤40 years old >40 years old

No. (%) No. (%) P value

Patients 29 (8.5%) 313 (91.5%) Males 15 (51.7%) 148 (47.3%) 0.647 Females 14 (48.3%) 165 (52.7%) Median age 33 (20-40) years 65 (41-91) years Smoking status Smoker 2 (6.9%) 70 (22.4%) 0.027 Non-smoker 14 (48.3%) 112 (35.8%) Ex-smoker 11 (37.9%) 70 (22.4%) Unknown 2 (6.9%) 61 (19.5%) Tumour grade Well differentiated 9 (31.0%) 85 (27.2%) 0.019 Moderate differentiated 19 (65.5%) 158 (50.5%) Poorly differentiated 0 43 (13.7%) Unknown 1 (3.5%) 27 (8.6%) TNM stage Stage I 8 (27.6%) 72 (23.0%) 0.922 Stage II 10 (34.5%) 113 (36.1%) Stage III 4 (13.8%) 55 (17.6%) Stage IV 7 (24.1%) 73 (23.3%)

Table 2. Pattern of recurrence stratified for age

Site of recurrence ≤40 years old >40 years old

Local 3 (10.3%) 38 (12.1%) 0.418

Regional 4 (13.8%) 39 (12.5%)

Distant 0 14 (4.5%)

Local and regional 2 (6.9%) 14 (4.5%)

Total 9 (31.0%) 105 (33.5%)

(10)

Table 3. Characteristics of the study group stratified for time period Characteristic <Year 2000 ≥Year 2000

Patients 94 (27.5%) 248 (72.5%) Males 47 (50.0%) 116 (46.7%) 0.594 Females 47 (50.0%) 132 (53.2%) Median age 64 (22-89) years 64 (20-91) years Smoking status Smoker 25 (26.6%) 47 (19.0%) 0.059 Non-smoker 27 (28.2%) 99 (39.9%) Ex-smoker 19 (20.2%) 62 (25.0%) Unknown 23 (24.5%) 40 (16.1%) Tumour grade Well differentiated 27 (28.2%) 67 (27.0%) 0.822 Moderate differentiated 46 (48.9%) 131 (52.8%) Poorly differentiated 14 (14.9%) 29 (11.7%) Unknown 7 (7.5%) 21 (8.5%) TNM stage Stage I 20 (21.3%) 60 (24.2%) 0.298 Stage II 33 (35.1%) 90 (36.3%) Stage III 22 (23.4%) 37 (14.9%) Stage IV 19 (20.2%) 61 (24.6%)

Table 4. Patterns of recurrence stratified for time period Site of recurrence <Year 2000 ≥Year 2000

Local 17 (18.1%) 24 (9.7%) 0.707

Regional 15 (16.0%) 28 (11.3%)

Distant 5 (5.3%) 9 (3.6%)

Total 41 (43.6%) 73 (29.4%)

(11)

3.2 Survival based on time period for diagnosis

Patients diagnosed after 2000 had an OS of 69.4% and 56.5% for 2 and 5 years respectively. For patients diagnosed before 2000 the OS was 46.8% at 2 years and 41.5% at 5 years. The survival difference was significant (p=0.004, figure 3). The general characteristics of the groups showed a borderline significance (p=0.059) on smoking status, the group diagnosed before 2000 having a higher rate of smokers but also a higher rate of patients with unknown smoking habits. There were no inter-group differences in gender, tumour differentiation grade, overall TNM stage or recurrence events (table 3, table 4).

3.3 Survival based on gender

Figure 4 compares survival based on patient gender. Outcomes for males and females were similar (p=0.949). Females were slightly older at diagnosis with a median age of 65 years (22-91) compared to males’ 62 years (20-88) and smoking habits differed significantly between the groups, males having a higher rate of smokers and

ex-Figure 2. Comparison on overall survival for the first 5 years after diagnosis between young (≤40 years) and old (>40 years) patients. Log-rank p=0.012. Numbers below the graph represent the number of patients still at risk per each 0.5 year.

Figure 3. Comparison on overall survival for the first 5 years after diagnosis for patients diagnosed before (<) or after (≥) the year 2000. Log-rank p=0.004. Numbers below the graph represent the number of patients still at risk per each 0.5 year.

(12)

differentiation grade and overall TNM stage. See table 5 for patient characteristics. There were no significant differences in recurrence patterns between the genders (table 6).

3.4 Survival based on TNM stage

When comparing survival based on TNM stage it followed an expected order as the highest survival rate was seen for stage I disease and the lowest for stage IV, with stage II and III in between. Log-rank p<0.001. See figure 5. When comparing the overall characteristics of the groups only patient gender showed some level of similarity across the groups. Median age was around 63 years for all four groups except the stage III group that had a median age of 71 years (26-91). Smoking habits, tumour differentiation grade and recurrence events all differed significantly between the groups (table 7, table 8).

4. Discussion

As mentioned earlier several studies have previously examined the relation between young age and survival in OTC. The studies have had no uniform but rather opposite conclusions [7,13,14,21–23], and the hypothesis that young tongue cancer have a lower survival than old has remained. The main objective of the study was therefore to investigate this hypothesis by studying the entire group of patients managed at our institution without matching cases with controls based on tumour differentiation grade, TNM stage and so forth. The results showed a considerably higher survival rate for the group of young adults, the difference being consistent over the entire 5-year follow-up period. Despite the small number of young patients, the survival difference was statistically significant (figure 2) while the general group characteristics

regarding gender, TNM stage and recurrence patterns were similar across groups (table 1, table 2).

The definition of a young tongue cancer patient varies and several different ages have been used as a cut-off. Most previous studies on the subject however have set this

(13)

point to 40 years [7,10,13,14,21,23], which is why this limit was used in the present study as well since this gives more comparable results.

In their report from 2011, Hammarstedt et al. [19] showed only a modest

improvement on survival for tongue cancer patients in Sweden during the years 1960-2004 compared to other oropharyngeal cancers. In the present study a similar

comparison on treatment success rate over time was made for patients treated at our institution. Since the study period included the years 1988-2014 year 2000 was used as a cut-off, being a good round number almost at the middle of the study period. Data showed that 5-year OS had improved by 36.1% during the study period (figure 3). Hammarstedt et al. have in their report divided the groups in decades and

according to gender. The observed survival rate for the years 1990-1999 was for men 41.0% (36.6-45.4) and for women 45.8% (40.5-51.0). For the years 2000-2004 the survival rates were 42.2% (35.9-48.2) and 52.4% (45.4-58.9) respectively. These numbers are fairly similar to the survival rate seen for the earlier time span in the present study with the observed survival rate of 41.5% being covered by the

confidence intervals above. For the later period, 2000-2014, the comparison is more difficult to make as Hammarstedts et al.’s report only includes the first five years of that period. The observed survival rate in the present study is well above the figures presented by Hammarstedt et al. although still within the CI presented for females. This seem logical given the assumption survival rates keep improving during the 21st century although it is difficult to estimate how this figure would translate for the entire Sweden. Several studies have previously reported a better outcome for female patients with OTC [13,14,19], or oral cavity cancer in general [8]. In the present study however, survival curves for both sexes were closely entwined and it was not possible to distinguish any significant survival benefit of female gender (figure 4). What this difference is due to is difficult to say. It is possible that the effect is caused by some matching criteria in the previous reports but it could also show a true population variance within Sweden where these results may have differed, if the study had been conducted elsewhere. One fact to support this theory is that a majority of the patients were women in the present study whereas Annertz et al. reported a M:F ratio of ~1.65 for ages 20-79 years during the period 1960-2008 [6,10]. One could argue that with the on-going gender empowerment in Sweden and elsewhere [26], with more and

(14)

and females, it is possible this will also be reflected by a more equal gender

distribution among patients with OTC. This however, is contradicted by the fact that Annertz et al. found the increasing incidence to be similar for both genders, even during the latest time period 2000-2008 [6]. Furthermore, since a not insignificant portion of the patients seem to lack these traditional risk factors it is possible that other, presently unknown, factors play a role in the aetiology of this disease.

4.1 Limitations

One important limitation of this study is that there is a varying delay in the reporting to the cancer registry. It generally tends to take a few weeks, but could potentially take considerably longer, from actual diagnosis or recurrence event until this information is reported to the registry. Because the registry only uses the date an event was reported and not when it was actually first discovered, time of death being and important exception, this means that all survival times in this study are slightly shorter than in reality. It is however reasonable to believe this time-lag being similar over the entire study population and thus should not have a major impact on the results.

Since tobacco smoking is one of the major risk factors for OTC [6] smoking status was included for patient group comparisons. It would have been favourable to also include data on alcohol consumption as alcohol seem to be a risk factor for oral cavity cancer on its own, but especially in combination with cigarette smoking [17]. But because of the retrospective art of the study, with access only to existing patient records, it was highly unlikely to find accurate information on alcohol intake for a large portion of the patients. It also turned out information on smoking habits was missing in the records for many patients, especially for the earlier years. The

information found was included anyway to give some idea of smoking status across groups although one should be cautious when interpreting these results because of the varying number of patients with unknown smoking habits, but also for the possibility that some of the information may be incorrectly noted in patient records. Patient were only categorized as being current smoker, previous smoker or non-smoker which is a very blunt method for categorizing tobacco smoke exposure compared to the much more describing pack-years. Again, this was due to the limited information in the

(15)

patient records. In the present study the rate of smoker were found to be higher among the OP but in addition to this, they also are very likely to have a much higher

cumulative exposure to tobacco smoke over time than the YP simply because they have lived longer. It is reasonable to believe that this affects the results at least on some level, as smoking is well known to shorten lives, but to what extent is difficult to say.

In the present study 5-year survival was the primary outcome used. Since this study contains no information on cause of death but only if patients were alive or diseased, the age difference when comparing YP and OP will inevitably affect the results. While the median age of the OP was only 65 years, the oldest patient in that group was 91 years at diagnosis. For obvious reasons an old patient have a higher

probability of dying by any other natural cause within five years than a young patient. To compensate for this in part, OS was specified for both 2-years and 5-years.

Studying the Kaplan-Meier survival curves (figures 2-5) it is also evident that the first two years after diagnosis consistently show the highest mortality rate in this study, which also corresponds to results seen in previous reports [13,19]. This was also the reason to include patients with as little as two years of follow-up to hopefully include a few more of the rare young patients.

It is noteworthy that no special consideration was taken regarding patients lost to follow-up. For the entire study group a total of ten patients (2.9%) were in one way or another missing standard follow-up protocol in the registry. This means data on recurrence events and possibly death could be missing on these patients, thus affecting the results.

5. Conclusion

Young age was associated with a significantly higher 5-year OS for patients with SCC of the oral tongue. Results of the present study do not support the hypothesis that young tongue cancer patients have a higher mortality than old. As for survival over time, the OS has improved during the study period, but only moderately, and survival

(16)

rate for patients with SCC of the oral tongue remain below 60 per cent. Further research is needed to optimize treatment and improve outcome for this patient group.

6. References

1. Eduardo M. Diaz J, Sturgis EM, Laramore GE, Sabichi AL, Lippman SM, Clayman G. Chapter 90. Neoplasms of the Head and Neck. In: Holland-Frei Cancer Medicine [Internet]. 6th edition. BC Decker Inc.; 2003 [cited 2017 Feb 22]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK12895/

2. Petersen PE. Strengthening the prevention of oral cancer: the WHO perspective. Community Dent Oral Epidemiol. 2005 Dec;33(6):397–9.

3. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012 Feb;62(1):10–29.

4. Liang X-H, Lewis J, Foote R, Smith D, Kademani D. Prevalence and significance of human papillomavirus in oral tongue cancer: the Mayo Clinic experience. J Oral Maxillofac Surg. 2008 Sep;66(9):1875–80.

5. Anatomy of Oral Tongue Cancer: HeadandNeckCancerGuide.org [Internet]. [cited 2017 Jun 5]. Available from:

http://www.headandneckcancerguide.org/adults/introduction-to-head-and-neck-cancer/oral-cancers/tongue-cancer/anatomy/

6. Annertz K, Anderson H, Palmér K, Wennerberg J. The increase in incidence of cancer of the tongue in the Nordic countries continues into the twenty-first century. Acta Otolaryngol. 2012 May;132(5):552–7.

7. Veness MJ, Morgan GJ, Sathiyaseelan Y, Gebski V. Anterior tongue cancer: age is not a predictor of outcome and should not alter treatment. ANZ J Surg. 2003 Nov;73(11):899–904.

8. Listl S, Jansen L, Stenzinger A, Freier K, Emrich K, Holleczek B, et al. Survival of patients with oral cavity cancer in Germany. PLoS ONE. 2013;8(1):e53415. doi:10.1371/journal.pone.0053415.

9. Salem A. Dismissing links between HPV and aggressive tongue cancer in young patients. Ann Oncol. 2010 Jan;21(1):13–7.

10. Annertz K, Anderson H, Biörklund A, Möller T, Kantola S, Mork J, et al.

Incidence and survival of squamous cell carcinoma of the tongue in Scandinavia, with special reference to young adults. Int J Cancer. 2002 Sep 1;101(1):95–9. 11. Müller S, Pan Y, Li R, Chi AC. Changing trends in oral squamous cell carcinoma

with particular reference to young patients: 1971-2006. The Emory University experience. Head Neck Pathol. 2008 Jun;2(2):60–6.

(17)

12. Shield KD, Ferlay J, Jemal A, Sankaranarayanan R, Chaturvedi AK, Bray F, et al. The global incidence of lip, oral cavity, and pharyngeal cancers by subsite in 2012. CA Cancer J Clin. 2017 Jan;67(1):51–64.

13. Blanchard P, Belkhir F, Temam S, El Khoury C, De Felice F, Casiraghi O, et al. Outcomes and prognostic factors for squamous cell carcinoma of the oral tongue in young adults: a single-institution case-matched analysis. Eur Arch

Otorhinolaryngol. 2017 Mar;274(3):1683–90.

14. Yip CSP, Charn TC, Wee JTS, Tan TWK, Goh C, Tan HK, et al. Outcomes of oral tongue cancer: does age matter? Ann Acad Med Singap. 2010

Dec;39(12):897–897.

15. Warnakulasuriya S. Global epidemiology of oral and oropharyngeal cancer. Oral Oncol. 2009 May;45(4–5):309–16.

16. The National Board of Health and Welfare [Internet]. [cited 2017 Apr 27]. Available from: http://www.socialstyrelsen.se/english

17. Toporcov TN, Znaor A, Zhang Z-F, Yu G-P, Winn DM, Wei Q, et al. Risk factors for head and neck cancer in young adults: a pooled analysis in the INHANCE consortium. Int J Epidemiol. 2015 Feb;44(1):169–85.

18. Hobbs CGL, Sterne J a. C, Bailey M, Heyderman RS, Birchall MA, Thomas SJ. Human papillomavirus and head and neck cancer: a systematic review and meta-analysis. Clin Otolaryngol. 2006 Aug;31(4):259–66.

19. Hammarstedt L, Lu Y, Marklund L, Dalianis T, Munck-Wikland E, Ye W. Differential survival trends for patients with tonsillar, base of tongue and tongue cancer in Sweden. Oral Oncol. 2011 Jul;47(7):636–41.

20. Rusthoven K, Ballonoff A, Raben D, Chen C. Poor prognosis in patients with stage I and II oral tongue squamous cell carcinoma. Cancer. 2008 Jan 15;112(2):345–51.

21. Hyam DM, Conway RC, Sathiyaseelan Y, Gebski V, Morgan GJ, Walker DM, et al. Tongue cancer: do patients younger than 40 do worse? Aust Dent J. 2003 Mar;48(1):50–4.

22. Hilly O, Shkedy Y, Hod R, Soudry E, Mizrachi A, Hamzany Y, et al. Carcinoma of the oral tongue in patients younger than 30 years: comparison with patients older than 60 years. Oral Oncol. 2013 Oct;49(10):987–90.

23. Liao C-T, Wang H-M, Hsieh L-L, Chang JT-C, Ng S-H, Hsueh C, et al. Higher distant failure in young age tongue cancer patients. Oral Oncol. 2006

Aug;42(7):718–25.

24. Park J-O, Sun D-I, Cho K-J, Joo Y-H, Yoo H-J, Kim M-S. Clinical outcome of squamous cell carcinoma of the tongue in young patients: a stage-matched comparative analysis. Clin Exp Otorhinolaryngol. 2010 Sep;3(3):161–5.

(18)

25. Garavello W, Spreafico R, Gaini RM. Oral tongue cancer in young patients: a matched analysis. Oral Oncol. 2007 Oct;43(9):894–7.

26. Hitchman SC, Fong GT. Gender empowerment and female-to-male smoking prevalence ratios. Bull World Health Organ. 2011 Mar 1;89(3):195–202.

(19)

7. Appendix

Table 5. Characteristics of the study group stratified for gender

Characteristic Males Females

Patients 163 (47.7%) 179 (52.3%) Median age 62 (20-88) years 65 (22-91) years Smoking status Smoker 42 (25.8%) 30 (16.8%) <0.001 Non-smoker 47 (28.8%) 79 (44.1%) Ex-smoker 52 (31.9%) 29 (16.2%) Unknown 22 (13.5%) 41 (22.9%) Tumour grade Well differentiated 43 (26.4%) 51 (28.5%) 0.400 Moderate differentiated 86 (52.8%) 91 (50.8%) Poorly differentiated 24 (14.7%) 19 (10.6%) Unknown 10 (6.1%) 18 (10.1%) TNM stage Stage I 31 (19.0%) 49 (27.4%) 0.111 Stage II 60 (36.8%) 63 (35.2%) Stage III 26 (16.0%) 33 (18.4%) Stage IV 46 (28.2%) 34 (19.0%)

Table 6. Patterns of recurrence stratified for gender

Site of recurrence Males Females

Local 21 (12.9%) 20 (11.2%) 0.865

Regional 20 (12.3%) 23 (12.8%)

Distant 8 (4.9%) 6 (3.4%)

Total 56 (34.4%) 58 (32.4%)

(20)

Figure 4. Comparison on overall survival by patient gender, for the first 5 years after diagnosis. Log-rank p=0.949. Numbers below the graph represent the number of patients still at risk per each 0.5 year.

Figure 5. Comparison on overall survival for the first 5 years after diagnosis by disease stage. Log-rank p<0.001. Numbers below the graph represent the number of patients still at risk per each 0.5 year.

(21)

T abl e 7. C ha ra ct er is ti cs of the s tudy gr oup st ra ti fi ed for T N M s ta ge C ha ra ct er is tic S ta ge I S ta ge II S ta ge II I S ta ge IV N o. ( % ) N o. ( % ) N o. ( % ) N o. ( % ) P va lue P at ie nt s 80 (23. 4% ) 123 (36 .0% ) 59 (17. 3% ) 80 (23. 4% ) Ma le s 31 (38. 8% ) 60 (48. 8% ) 26 (44. 1% ) 46 (57. 5% ) 0. 111 Fe m al es 49 (61. 3% ) 63 (51. 2% ) 33 (55. 9% ) 34 (42. 5% ) M edi an age 63 (28 -90) ye ar s 63 (22 -91) ye ar s 71 (26 -91) ye ar s 62 .5 (20 -89) ye ar s S m oki ng st at us S m oke r 10 (12 .5 %) 24 (19 .5 %) 13 (22 .0 %) 25 (31 .3 %) 0. 012 N on -s m oke r 38 (47 .5 %) 51 (41 .5 %) 16 (27 .1 %) 21 (26 .3 %) Ex -s m oke r 24 (30 .0 %) 24 (19 .5 %) 16 (27 .1 %) 17 (21 .3 %) U nknow n 8 (10 .0 %) 24 (19 .5 %) 14 (23 .7 %) 17 (21 .3 %) T um our gr ade W el l di ff er ent ia te d 29 (36 .3 %) 38 (30 .9% ) 15 (25 .4 %) 12 (15 .0 %) <0 .001 M ode ra te di ff er ent ia te d 41 (51 .3 %) 59 (48. 0%) 34 (57 .6 %) 43 (53 .8 %) P oor ly di ff er ent ia te d 3 (3 .8 %) 12 (9 .8 %) 5 (8 .5 %) 23 (28 .8 %) U nknow n 7 (8 .8 %) 14 (11 .4 %) 5 (8 .5 %) 2 (2 .5 %) Tabl e 8. P at te rns of r ec ur re nc e st ra ti fi ed for T N M s ta ge S it e of r ec ur re nc e S ta ge I S ta ge II S ta ge II I S ta ge IV N o. ( % ) N o. ( % ) N o. ( % ) N o. ( % ) P va lue L oc al 12 (15. 0%) 8 (6. 5%) 9 (15. 3%) 12 (15. 0%) 0. 003 R egi ona l 12 (15. 0%) 18 (14. 6%) 5 (8. 5%) 8 (10. 0%) D is ta nt 1 (1. 3%) 4 (3. 3%) 1 (1. 7%) 8 (10. 0%) L oc al a nd re gi ona l 0 7 (5. 7%) 6 (10. 2%) 8 (3. 8%) T ot al 25 (31. 3%) 37 (30. 1%) 21 (35. 6%) 31 (38. 8%)

(22)

Cover letter

May 19th, 2017.

Dear Editor

I am writing to submit the manuscript entitled “Prognostic factors in oral tongue

cancer – does age, gender or time period of treatment predict outcome?” for

consideration for publication.

This retrospective study has investigated the overall survival rates for patients with squamous cell carcinoma of the oral tongue to see if young age at diagnosis (≤40 years) predicts to worse survival compared to older individuals (>40 years). Similar studies have been conducted previously but with conflicting results and data has never been published on a Scandinavian population.

The study population includes all tongue cancer patients treated at Örebro University Hospital during the years 1988-2014. Young age was associated with a considerably higher 5-year overall survival, the difference being statistically significant. I feel that it might be of interest to your readers.

This study has not been published before and is not considered for publication elsewhere.

Sincerely Yours,

Oscar Smedberg (corresponding author), Bachelor of Medicine, Örebro University.

(23)

Populärvetenskaplig sammanfattning

Tungcancer är en gradvis ökande men fortfarande relativt ovanlig cancersjukdom i Sverige. Rökning och hög alkoholkonsumtion är de främsta kända riskfaktorerna för denna sjukdom som främst drabbar personer i övre medelåldern. Män drabbas generellt lite oftare än kvinnor.

Behandling av tungcancer innefattar ofta en kombination av kirurgi och strålbehandling och kan leda till stort lidande för patienten. Trots aggressiva behandlingsinsatser är prognosen relativt dålig jämfört med flera andra

cancersjukdomar och den förväntade överlevnaden i tungcancer varierar från 70 procent ned mot 15 procent beroende på hur snabbt den upptäcks.

I en ny studie från Örebro Universitet har forskare undersökt sambandet mellan insjuknandeålder och chansen att överleva sjukdomen på fem års sikt, ett mått som vanligtvis används för att beskriva överlevnaden i tumörsjukdomar. Studien

omfattade 342 patienter och det visade sig att en tiondel av patienterna var under 40 år när de fick sin diagnos. Dessa hade 50 procent bättre chans att överleva än gruppen patienter som var över 40 år. Det var en viss förbättrad överlevnad för patienter diagnostiserade mellan åren 2000 till 2014 jämfört med dem som fått sin diagnos mellan 1988 och 1999. Man såg ingen skillnad i överlevnad mellan män och kvinnor.

Etiskt övervägande

Den här studien är gjord i form av ett kvalitetsarbete på ÖNH-kliniken och baseras på tidigare insamlad data i journaler och ett regionalt cancerregister. I egenskap av viktigt kvalitetsarbete för klinikens räkning fanns givetvis ett godkännande från verksamhetschefen att gå igenom aktuella journaler. Studiepopulationen utgörs av både levande och avlidna patienter och studien har gjorts utan att några patienter tillfrågats. Sekretessfrågan är, som alltid, viktig inom hälso- och sjukvården och jag har varit mån om att all datahantering skötts korrekt och att inga uppgifter lämnat kliniken innan de varit avidentifierade. Journalgranskningen har också skett så systematiskt som möjligt för att efterleva det dokument med etiska riktlinjer

(24)

Lars-Tanken var att söka etikgodkännande för denna studie för att öppna möjligheten att publicera resultaten men tyvärr fick vi konstatera att vi inte skulle hinna få ett svar inom rimlig tid. Det är synd med tanke på att den integritetskränkning forskningen ändå innebär får komma till sin rätt först när resultaten görs tillgängliga för andra. Förhoppningsvis kan resultaten i alla fall spridas här på kliniken i Örebro så att kunskapen ändå till viss del kommer till nytta. Man brukar också såga att forskning ska vara till nytta för patienten vilket är svårt att säga att det är i en retrospektiv studie som denna. Dock kan man tänka sig indirekt nytta om t ex en familjemedlem senare skulle drabbas och kunskapen då kan komma till användning.

Prognostic factors in oral tonguecancer – does age, gender or timeperiod of treatment predictoutcome?