Oral Cavity Squamous Cell Carcinoma in Young Patients: A Multi-Institutional Study of the Canadian Head & Neck Collaborative Research Initiative

Abstract

Importance

Oral cavity squamous cell carcinoma (OCSCC) is rare in patients ≤40 years, and their risk factors, presentation, and outcomes may differ from older patients.

Objective

To assess the epidemiology, risk factors, and oncologic outcomes of young patients (≤40 years) with OCSCC compared to those >40 years.

Design

A multi-institutional retrospective cohort study.

Setting

Nine Canadian institutions from 2005 to 2019.

Participants

In total, 4506 adults with OCSCC, of whom 205 (4.55%) were young and 4301 were older than 40.

Interventions or Exposures

The primary outcomes were overall survival (OS) and disease-free survival (DFS), comparing young and older patients. The identification of risk factors for OCSCC development in young patients was a secondary outcome.

Main Outcome Measures

OS, DFS, and risk factor identification.

Results

Oral tongue cancer was the most common subsite (48.9%), with a significantly higher proportion of cases in young patients (73.4% vs 47.7%, P < .01). Young patients were more likely to present at an earlier clinical stage (T1: 44% young vs 31% old, P < .01) and were less likely to smoke (57% young vs 31% old, P < .01) or consume alcohol (72% young vs 58% old, P < .01). Multivariable analysis showed that smoking status, previous head and neck cancer, and advanced stage were significantly associated with decreased OS and DFS (P < .05). No significant differences were found in local (P = .61), regional (P = .67), or distant (P = .50) disease failure between age groups.

Conclusions and Relevance

Young patients with OCSCC were less likely to smoke or drink and presented at earlier stages, but they did not experience improved OS or DFS compared to older patients. These findings emphasize the need for further research into biological differences in OCSCC between young and older patients.

Keywords

oral cavity head and neck oropharynx head and neck (malignancy)pediatric general pediatrics

Graphical Abstract

Key Points

Young patients with OCSCC (≤40 years) are less likely to exhibit traditional risk factors such as smoking and alcohol consumption, yet they often present with earlier-stage disease compared to older patients.

Despite earlier presentation, young patients do not show improved overall survival (OS) or disease-free survival (DFS) outcomes compared to those over 40 years, suggesting potential differences in tumor biology.

Further research is necessary to explore the biological differences in OCSCC between younger and older patients to better understand the underlying factors influencing outcomes.

Introduction

Head and neck cancers account for 4% of all cancers in the United States and Canada, with oral cavity squamous cell carcinoma (OCSCC) being the most common.^1,2 According to statistics gathered in 2017, the age-standardized incidence of OCSCC was 11.9 cases per 100,000 and the overall 5-year survival was 63%.² The most common etiologic agents for OCSCC are tobacco and alcohol,³ which are known to increase OCSCC risk in a synergistic manner.⁴ Evidence suggests that longer durations, rather than greater intensity of smoking and alcohol use, predict the development of OCSCC.^5,6 Consequently, the risk of developing OCSCC increases with age.⁷

Despite the known association between OCSCC development and age, smoking and alcohol use, there is a small subset of OCSCC that is observed among young, non-smokers and non-drinkers for which the incidence of OCSCC is rising.^8
-11 Even among young smokers and drinkers, the total duration of tobacco and alcohol consumption seems insufficient to explain OCSCC development. While the epidemiology, biology, treatment, and prognosis of OCSCC in patients with the typical risk factors of chronic smoking and alcohol exposure have been well studied and reported, those of OCSCC in young patients remain largely unknown and controversial.

Previous small observational studies have demonstrated that young patients with OCSCC tend to present with earlier stage disease but suffer from clinically aggressive disease behavior and may have more frequent recurrences.^11
-16 Thus far, the literature investigating age as it relates to survival and oncologic outcomes in OCSCC presents conflicting evidence, with studies yielding differing conclusions on the potential protective, harmful, or neutral effect of age on overall and disease-free survival.^12,17,18 The discrepancies in such findings may be explained by small sample sizes resulting in underpowered studies, as well as potential genetic differences across different geographic populations.^11,12 Such discrepancies in reported outcomes within this patient population make forming decisions around treatment and predicting prognosis challenging.¹⁹

To reconcile these conflicting findings, we performed a multi-institutional national retrospective cohort study to describe the epidemiology and oncologic outcomes of OCSCC in young patients.

Methodology

Participating Centers

We completed a multicenter, retrospective analysis of OCSCC patients through the Canadian Head & Neck Collaborative Research Initiative, a national network of surgical oncologists. Each center obtained local ethics approval from its respective institutional review boards, and data were shared and combined for analysis through data use agreements. Individual informed consent was waived at all centers owing to use of a limited, deidentified data set.

Study Population

All surgeons were invited to contribute their data if they were able to provide cases for the study period, that is, between January 1, 2005 and December 31, 2019. Surgeons contributing to the data set were from 9 participating centers in Canada. Inclusion criteria were as follows: (1) adults aged 18 and above, (2) OCSCC (any subsite) diagnosed during the study period, and (3) diagnosed or treated at a participating institution. The exclusion criteria were as follows: (1) carcinoma in situ, (2) tumors of the oropharynx, (3) cutaneous lip squamous cell carcinoma (4) tumor pathologies of the oral cavity other than squamous cell carcinoma, and (5) a lack of data available on exposure and outcome variables. Patients were subdivided based on being ≤40 years of age (“young”) or > 40 years of age (“old”) at the time of OCSCC diagnosis. The terms “young” and “old” will be used in this article strictly for convenience in distinguishing the cohorts.

Data Collection and Sources

A standardized extraction form was developed. Study data were collected and managed on password-protected Microsoft Excel datasheets. Study data collected included demographic, clinical, and histopathologic variables as well as treatment and survival outcomes. A list of all variables collected and their definitions is detailed in Appendix I.

Outcomes

Primary outcomes included overall survival (OS) and disease-free survival (DFS) in young versus old OCSCC patients. OS was defined as the length of time from the date of diagnosis to the date of death. DFS was defined as the length of time from the date of diagnosis to the date of local, regional, or distant recurrence in a living patient. Secondary outcomes included differences in risk factors for OCSCC development between young and old patients.

Statistical Analysis

Descriptive analysis was performed to assess patient characteristics. Continuous variables were expressed as medians and interquartile ranges (IQR). Crude comparisons among patients were performed using the Mann-Whitney U test. Categorical variables were expressed as frequencies with 95% confidence intervals (CIs) and were compared using the chi-square test. The overall survival (OS) and disease-free survival (DFS) rates were computed from the survival curves estimated by the Kaplan-Meier method, and groups were compared using the log-rank test. Hazard ratios (HR) and 95% CIs were calculated. Factors that were either statistically significant on univariable analysis or felt to be clinically important were included in multivariable analysis using Cox proportional hazards modelling. In all analyses, a two-sided P < .05 was considered statistically significant.

Results

Baseline Demographic Characteristics

In total, 4506 patients from 9 institutes were included in the study. At the time of OCSCC diagnosis, 205 patients (4.55%) were ≤40 years of age (“young”), and 4301 patients (95.45%) were >40 years of age (“old”). Patient baseline demographic characteristics are summarized in Table 1. The median age at diagnosis was 35 (IQR = 31-39) and 65 (IQR = 57-74) years amongst young and old patients, respectively. Overall, 1735 patients (38.50%) were female, and this proportion was not significantly different between the young (41.46%) and old groups (38.36%). Among young patients, 90.48% had an Eastern Cooperative Oncology Group (ECOG) performance status of 0, whereas only 65.61% of old patients had a performance status of 0 (P < .01).

Table 1.

Baseline Characteristics (Age ≤40).

Characteristic	Young (≤40 years old)n (%)	Old (>40 years old)n (%)	P-value
N	205	4301
Age at diagnosisMedian (IQR)	35 (31-39)	65 (57-74)	<.01
Female sex	85 (41.46)	1650 (38.36)	.38
ECOG
0	114 (90.48)	1986 (65.61)	<.01
1	12 (9.52)	785 (25.93)
2-4	0 (0)	256 (8.46)
Smoking status			<.01
Never	109 (54.77)	1307 (31.04)
Previous	44 (22.11)	1319 (31.32)
Current	46 (23.12)	1585 (37.64)
Quantity if current or previous (pack yrs) median (IQR)	13.27 (4-20)	35 (20-50)	<.01
Alcohol status			<.01
Never	137 (71.73)	2400 (58.37)
Previous	14 (7.33)	554 (13.47)
Current	40 (20.94)	1158 (28.16)
Quantity if current or previous (drinks/week) Median (IQR)	20 (12-30)	21 (15-35)	.32
Charleson-Deyo Comorbidity Index			<.01
0-2	69 (69.70)	854 (35.13)
3-5	4 (4.04)	831 (34.18)
6+	26 (26.26)	746 (30.69)
Immunosuppression			.33
Yes	4 (3.48)	62 (2.18)
No	111 (96.52)	2781 (97.82)
Type of immunosuppression			.13
Solid organ transplant	0 (0)	12 (20)
Bone marrow transplant	2 (50)	6 (10)
HIV	0 (0)	15 (25)
Other	2 (50)	27 (45)
Previous head and neck cancer			<.01
Yes	4 (1.98)	280 (6.53)
No	198 (98.02)	4008 (93.47)
Previous non-head and neck cancer			.03
Yes	7 (4.43)	350 (9.66)
No	151 (95.57)	3272 (90.34)

IQR, interquartile range.

The proportion of young patients who never smoked was substantially higher than among older patients (54.77% vs 31.04%, P < .01). Additionally, 37.64% of older patients continued to smoke at the time of diagnosis, compared to 23.12% of young patients (P < .01). Old patients demonstrated greater overall tobacco use compared to young patients (median smoking pack-year history 35.00 (IQR = 20-50) vs 13.27 (IQR = 4-20), respectively, P < .01). A larger proportion of young patients had never consumed alcohol (71.73%), as compared to old patients (58.37%) (P < .01), but the median number of drinks per week was similar between both groups at 20 (IQR 12-30) for young, and 21 (IQR 15-35) for old, respectively (P = .32). The proportion of patients with a Charlson-Deyo Comorbidity Index (CDCI) of 0-2 for young and old patients was 69.70% and 35.13%, respectively (P < .01). The proportion of patients with a CDCI >6 for young and old patients was 26.26% and 30.69%, respectively (P < .01). A greater number of young patients were immunosuppressed (3.48%) compared to old patients (2.18%); however, the difference was not statistically significant (P = .32).

OCSCC Characteristics

Oncologic characteristics of OCSCC cases are summarized in Table 2. The most common oral cavity subsite involved in young patients was the oral tongue (73.4%), followed by the buccal mucosa (6.9%), and the floor of mouth (4.4%). The most common oral cavity subsite involved in old patients was the oral tongue (47.7%), followed by the floor of mouth (17.3%), and the mandibular alveolus (10.5%). The majority of tumors in both cohorts were moderately differentiated (young 47.7%, old 57.7%). Among young patients, the majority of tumors were staged T1 (44.0%), whereas T2 tumors were most common among old patients (33.1%) (P < .01). At the time of presentation, the majority of patients in both cohorts did not have nodal disease (young 68.6% N0, old 67.8% N0) or distant metastasis (young 99.5% M0; old 98.4% M0). Old patients presented with significantly larger tumors (median tumor dimension 2 cm (IQR = 1.2-3.0 cm) among young vs 2.3cm (IQR=1.4-3.5 cm) among old, P = .04). Tumor thickness was similar between groups (median tumor thickness 7 mm (IQR 2-14 mm) among young, 7 mm (IQR=3-13 mm) among old) as was depth of invasion (median DOI 7.4 mm (IQR=4-13 mm) among young, 7 mm (IQR = 3-13 mm) among old). A greater proportion of old patients presented with tumor bony invasion than young patients (17.0% vs 7.3%, P < .01) while perineural invasion was more common in young compared to old patients (39.9% vs 32.6%, P < .05).

Table 2.

Cancer Characteristics (Age ≤40).

Characteristic	Young (≤40 years old)n (%)	Old (>40 years old)n (%)	P-value
Oral cavity subsite
Oral tongue	149 (73.40)	2051 (47.74)	<.01
Floor of mouth	9 (4.43)	743 (17.30)
Buccal mucosa	14 (6.90)	375 (8.73)
Retromolar trigone	5 (2.46)	257 (5.98)
Hard palate	6 (2.96)	105 (2.44)
Mucosal lip	8 (3.94)	142 (3.31)
Mandibular alveolus	8 (3.94)	449 (10.45)
Maxillary alveolus	4 (1.97)	174 (4.05)
Grade			.05
Well-differentiated	55 (35.48)	954 (29.49)
Moderately differentiated	74 (47.74)	1867 (57.71)
Poor differentiated	26 (16.77)	414 (12.80)
Clinical stage (T)			<.01
1	88 (44)	1319 (31.23)
2	70 (35)	1400 (33.14)
3	13 (6.50)	483 (11.43)
4	29 (14.50)	1022 (24.20)
Clinical stage (N)			.98
0	140 (68.63)	2881 (67.82)
1	26 (12.75)	521 (12.26)
2	36 (17.65)	799 (18.81)
3	2 (0.98)	47 (1.11)
Clinical stage (M)			.38
0	195 (99.49)	4057 (98.38)
1	1 (0.51)	67 (1.62)
Pathological stage (T)			.23
1	87 (46.28)	1464 (40.19)
2	58 (30.85)	1123 (30.83)
3	16 (8.51)	343 (9.42)
4	27 (14.36)	713 (19.57)
Tumor dimensions (cm) Median (IQR)	2 (1.2-3.0)	2.3 (1.4-3.5)	.04
Tumor thickness (mm) Median (IQR)	7 (2-14)	7 (3-13)	.81
Depth of invasion (mm) Median (IQR)	7.4 (4-13)	7 (3-13)	.55
Positive bone invasion	10 (7.30)	522 (17.03)	<.01
Positive perineural invasion	67 (39.88)	1088 (32.59)	.05
Positive lymphovascular invasion	29 (17.26)	548 (16.35)	.76
Margin			.59
Positive	35 (25.74)	807 (29.66)
Close	50 (36.76)	977 (35.91)
Clear	51 (37.50)	937 (34.44)
Pathological stage (N)			.13
0	67 (55.83)	1455 (59.53)
1	15 (12.50)	317 (12.97)
2	27 (22.50)	563 (23.04)
3	11 (9.17)	109 (4.46)
Size of largest metastatic deposit (cm) Median (IQR)	1.6 (0.9-2.6)	`.35 (0.3-2.5)	.33
Positive extranodal extension	27 (33.33)	495 (32.80)	.93

Treatment modalities

Treatment characteristics are shown in Table 3. The rate of curative intent treatment was higher among young than old patients (99.0% vs 93.0%, P < .05). A greater percentage of young patients underwent primary surgery compared to old patients (97.1% vs 88.3%, P < .05). Similar proportions of young and old patients underwent adjuvant radiation therapy (40.6% among young, 44.4% among old, P = .28); however, young patients were more likely to receive concurrent adjuvant chemotherapy (21.4% among young vs 12.8% among old, P < .05). The rate of neck dissections was similar among young and old patients (70.4% among young, 64.6% among old, P = .12). Young patients were followed for a median of 36.5 months (IQR = 14-62) from the time of diagnosis until the last documented follow-up, which was significantly greater than for old patients, who were followed for a median of 30 months (IQR = 11-56) (P < .05).

Table 3.

Treatment Modalities (Age ≤40).

Characteristic	Young (≤40 years old)n (%)	Old (>40 years old)n (%)	P-value
Treatment intent
Curative	201 (99.01)	3926 (93.03)	<0.01
Palliative	2 (0.99)	294 (6.97)
Surgery—primary tumor	199 (97.07)	3776 (88.31)	<0.01
Surgery—neck dissection	121 (70.35)	2507 (64.63)	0.13
Radiation	82 (40.59)	1893 (44.43)	0.29
Chemotherapy	43 (21.39)	542 (12.80)	<0.01
Months of follow-up from diagnosisMedian (IQR)	36.5 (14-62)	30 (11-56)	0.02

Clinical Outcomes

Clinical outcomes are shown in Table 4. Among young patients, 82.0% were alive at the time of study completion, compared to 70.6% of old patients (P < .05). However, of those who had died, 94.1% of young patients died from their disease, a significantly higher proportion than that of old patients (71.01%, P < .05). The median time from diagnosis to death among young and old patients was 18 (IQR = 10-26) and 14 (IQR = 7-27) months, respectively (P = .19). The OS as estimated by the Kaplan-Meier method demonstrated significantly (P < .05) worse OS in old compared to young patients (Figure 1). Young patients had local, regional, and distant failure rates of 17.0%, 18.9%, and 11.6%, respectively. Old patients had similar local, regional, and distant failure rates of 18.6%, 17.0%, and 10.0% (P = .60, P = .67, P = .50 respectively). The time from diagnosis to local, regional, and distant failure did not significantly differ across age groups. For local failure, young patients had a median time to local failure of 12.5 months (IQR = 6.5-25.5) compared to old patients with a median of 14 months (IQR = 7-34) (P = .44). Similarly, with regional failure, young patients had a median time of 11 months (IQR = 6-26), while old patients had a median time of 10 months (IQR = 6-19) (P = 0.52). Finally, for distant failure, the young age group had a median failure time of 19 months (IQR = 14-33), while the old group was 17 months (IQR = 10-27.5) (P = 0.78). Taken together, the DFS between young and old patients demonstrated no significant differences as estimated by the Kaplan-Meier method (Figures 2 -4).

Table 4.

Clinical Outcomes (Age ≤40).

Characteristic	Young (≤40 years old)	Old (>40 years old)	P-value
Alive, n (%)	168 (81.95)	3024 (70.62)	<0.01
Dead, n (%)	37 (18.05)	1258 (29.38)
Dead from disease	32 (94.12)	774 (71.01)	<0.01
Dead from other causes	2 (5.88)	316 (28.99)	-
Months from diagnosis to deathMedian (IQR)	18 (10-26)	14 (7-27)	0.19
Local failure, n (%)	28 (16.97)	673 (18.57)	0.61
Months from diagnosis to local failureMedian (IQR)	12.5 (6.5-25.5)	14 (7-34)	0.45
Regional failure, n (%)	32 (18.93)	643 (17.67)	0.68
Months from diagnosis to regional failureMedian (IQR)	11 (6-26)	10 (6-19)	0.52
Distant failure, n (%)	19 (11.59)	358 (9.95)	0.50
Months from diagnosis to distant failureMedian (IQR)	19 (14-33)	17 (10-27.5)	0.78

Figure 1.

Kaplan-Meier curve (unadjusted) for time from diagnosis to last follow-up in young versus old OCSCC patients.

Figure 2.

Kaplan-Meier curve (unadjusted) for time from diagnosis to local failure in young versus old OCSCC patients.

Figure 3.

Kaplan-Meier curve (unadjusted) for time from diagnosis to regional failure in young versus old OCSCC patients.

Figure 4.

Kaplan-Meier curve (unadjusted) for time from diagnosis to distant failure in young versus old OCSCC patients.

Overall Survival

The results of our uni- and multivariable analyses are depicted in Tables 5 and 6, respectively. Multivariable analysis revealed that OS did not significantly differ between young versus old age groups (HR 0.94, P = 0.78). Factors found to have significantly (P < 0.05) decreased OS were previous head and neck cancer (HR 1.42), clinical tumor stage compared to cT1 including cT2 (HR 1.71), cT3 (HR 1.95), cT4 (HR 1.86), nodal staging compared to cN0 including cN1 (HR 1.51), cN2 (HR 1.91), cN3 (HR 2.24), and distant metastatic disease (HR 1.45), as well as treatment modalities compared to surgery only including no treatment (HR 3.81), chemotherapy only (HR 2.87), and radiation only (HR 2.21).

Table 5.

Univariate Analysis (Age ≤40) of Overall Survival and Disease-Free Survival.

	Overall survival		Disease-free survival
	HR(95% CI)	P-value	HR(95% CI)	P-value
Baseline characteristics
Age (Old)	1.8(1.051-2.905)	.03	1.6(1.077-2.343)	.02
Sex (female)	0.9(0.824-1.043)	.21	0.9(0.814-0.994)	.04
Young female	0.5(0.359-0.560)	<.01	0.5(0.408-0.592)	<.01
ECOG (1)	1.8(1.320-2.479)	<.01	1.3(1.215-1.463)	<.01
ECOG (2-4)	3.7(2.576-5.320)	<.01	2.5(2.033-3.156)	<.01
Smoking status (previous)	1.3(1.144-1.573)	<.01	1.2(1.009-1.389)	.04
Smoking status (current)	1.3(1.089-1.502)	<.01	1.1(0.974-1.247)	.13
Smoking pack years	1.0(1.002-1.008)	<.01	1.0(1.001-1.005)	<.01
Alcohol status (previous)	1.4(1.129-1.624)	<.01	1.2(1.033 - 1.304)	.02
Alcohol status (current)	1.3(1.104-1.603)	<.01	1.2(1.069-1.287)	<.01
Immunosuppression characteristics
Immunosuppression (yes)	1.0(0.460-1.969)	.89	1.0(0.711-1.345)	.89
Solid organ transplant	2.4(0.845-6.725)	.10	1.9(0.764-4.564)	.17
Bone marrow transplant	0.4(0.164-0.859)	.02	0.5(0.205-1.009)	.05
HIV	0.7(0.277 -1.762)	.45	1.1(0.820-1.486)	.52
Other	0.8(0.354-1.597)	.46	0.7(0.484-1.085)	.12
Cancer characteristics
Previous head and neck cancer	1.4(1.271-1.585)	<.01	1.5(1.361-1.747)	<.01
Previous non-head and neck cancer	1.1(0.946-1.367)	.17	1.1(0.926-1.362)	.24
Floor of mouth	1.2(1.028-1.445)	.03	1.2(1.029-1.293)	.02
Buccal	1.1(0.904-1.331)	.35	1.3(1.153-1.457)	<.01
Retromolar trigone	1.3(0.926 - 1.833)	.13	1.2(1.010-1.513)	.04
Hard palate	1.5(0.819-2.692)	.19	1.2(0.697-2.116)	.50
Mucosal lip	0.6(0.372-1.124)	.12	0.8(0.466-1.283)	.32
Mandibular alveolus	1.1(0.735-1.782)	.55	1.2(0.860-1.553)	.34
Maxillary alveolus	0.9(0.673-1.279)	.65	1.0(0.760-1.212)	.73
Histologic grade (moderately differentiated)	1.6(1.439-1.836)	<.01	1.4(1.265-1.561)	<.01
Histologic grade (poorly differentiated)	2.4(2.025-2.843)	<.01	2.1(1.973-2.188)	<.01
Clinical stage
cT2	2.0(1.747-2.237)	<.01	1.7(1.493-1.988)	<.01
cT3	3.1(2.605-3.747)	<.01	2.45(2.128-2.825)	<.01
cT4	3.4(2.842-4.132	<.01	2.6(2.243-2.997)	<.01
cN1	2.0(1.604-2.522)	<.01	1.7(1.460-2.072)	<.01
cN2	2.8(2.296-3.360)	<.01	2.2(1.798-2.739)	<.01
cN3	3.3(1.189-8.862)	.02	2.7(1.245-5.831)	.02
cM1	3.7(2.641-5.266)	<.01	3.4(2.895 - 3.944)	<.01
Pathologic stage
pT2	2.1(1.690-2.610)	<.01	1.7(1.348-2.118)	<.01
pT3	2.5(2.132-2.966)	<.01	2.0(1.673-2.288)	<.01
pT4	2.9(2.350-3.580)	<.01	2.2(1.926-2.548)	<.01
pN1	1.8(1.534-2.070)	<.01	1.6(1.374-1.750)	<.01
pN2	3.4(2.829-4.000)	<.01	2.6(2.202-3.017)	<.01
pN3	4.7(3.005-7.311)	<.01	3.5(2.779-4.335)	<.01
Treatment group versus surgery only
No treatment	8.8(5.718-13.554)	<.01	5.2(3.766-7.063)	<.01
Chemotherapy only	7.6(4.373-13.360)	<.01	4.2(2.556-6.969)	<.01
Radiation only	4.0(3.528-4.464)	<.01	3.1(2.757-3.577)	<.01
Radiation + chemotherapy	2.2(1.718-2.915)	<.01	2.3(2.060-2.637)	<.01
Surgery + radiation + chemotherapy	1.8(1.468-2.259)	<.01	1.4(1.205-1.708)	<.01
Surgery + Radiation	1.3(1.133-1.547)	<.01	1.1(0.958-1.210)	.22
Surgery + chemotherapy	2.5(1.729-3.474)	<.01	2.9(2.017-4.063)	<.01

Table 6.

Multivariate analysis (Age ≤40) of Overall Survival and Disease-Free Survival.

	Overall survival		Disease-Free Survival
	HR(95% CI)	P-value	HR(95% CI)	P-value
Baseline characteristics
Age (Old)	0.9(0.609-1.448)	.78	1.0(0.663-1.555)	.95
Sex (female)	1.0(0.873-1.163)	.92	1.0(0.875-1.050)	.37
Smoking status (previous)	1.1(0.932-1.316)	.25	1.0(0.913-1.132)	.077
Smoking status (current)	1.0(0.752-1.191)	.64	0.8(0.762-0.932)	<.01
Alcohol status (previous)	1.0(0.757-1.266)	.88	1.0(0.825-1.132)	.680
Alcohol status (current)	1.0(0.793-1.234)	.93	1.1(0.913 - 1.208)	.49
Previous head and neck cancer	1.4(1.162-1.738)	<.01	1.5(1.053-1.998)	.03
Clinical stage
cT2	1.7(1.451-2.020)	<.01	1.7(1.461-1.880)	<.01
cT3	2.0(1.486-2.571)	<.01	1.9(1.499-2.299)	<.01
cT4	1.9(1.482-2.329)	<.01	1.8(1.479-2.254)	<.01
cN1	1.5(1.086-2.096)	.02	1.4(1.082-1.736)	<.01
cN2	1.9(1.372-2.646)	<.01	1.6(1.179 - 2.292)	<.01
cN3	2.2(0.973-5.142)	.06	2.1(1.094-3.984)	.03
cM1	1.5(1.012-2.067)	.05	1.5(1.028-2.161)	.04
Treatment group versus surgery only
No treatment	3.8(2.079-6.997)	<.01	2.2(1.373-3.356)	<.01
Chemotherapy only	2.9(2.103-3.922)	<.01	1.7(1.154-2.418)	<.01
Radiation only	2.2(1.810-2.694)	<.01	1.8(1.423-2.155)	<.01
Radiation + chemotherapy	1.1(0.746-1.559)	.69	1.1(0.829-1.429)	.54
Surgery + radiation + chemotherapy	1.0(0.642-1.541)	.99	0.8(0.549-1.155)	.23
Surgery + radiation	0.9(0.710-1.175)	.48	0.7(0.594-0.839)	<.01
Surgery + chemotherapy	1.5(0.761-3.104)	.23	1.7(0.788-3.578)	.18

Disease-Free Survival

DFS outcomes from univariable analysis and multivariable analysis are detailed in Tables 5 and 6, and summarized in Table 7 respectively. In multivariable analysis, young age did not carry a significantly reduced hazard of DFS (HR 1.02). Factors that were associated with decreased DFS included previous head and neck cancer (HR 1.45), clinical tumor stage compared to cT1 including cT2 (HR 1.66), cT3 (HR 1.86), cT4 (HR 1.83), clinical nodal staging compared to cN0 including cN1 (HR 1.37), cN2 (HR 1.64), cN3 (HR 2.09) distant metastatic disease (HR 1.49), and treatment modalities compared to surgery only including no treatment (HR 2.15), chemotherapy only (HR 1.67), and radiation only (HR 1.75).

Table 7.

Summary of Sub-Group Analysis for Patients with Stage I and II Disease.

	Young (age ≤40)	Old (age >40)	P-value
Total (% of total)	122 (5.4)	2128 (94.6)	-
Survival
Alive (%)	110 (90.16)	1706 (80.47)	<.01
Dead (%)
Dead from disease (% of age)	10 (90.91)	217 (62.18)	.06
Dead from other causes (% of age)	1 (9.09)	132 (37.82)
Treatment failure (Months)
Local failureMedian (IQR)	17 (9.5-33.5)	19 (8-47)	.74
Regional failureMedian (IQR)	18 (9-44)	12 (7-24)	.14
Distant failureMedian (IQR)	18 (18-19)	18 (10-28)	.87
Age risk factor analysis (Old >40)
OS Univariate analysisHR (95% CI)	1.9 (0.825-4.491)		.13
OS Multivariate analysisHR (95% CI)	0.9 (0.501-1.721)		.82
DFS Univariate analysisHR (95% CI)	1.9 (1.253-2.841)		<.01
DFS Multivariate analysisHR (95% CI)	1.3 (0.854-1.905)		.24

Discussion

In the last several decades, there has been a rise in OCSCC among young patients globally, and yet the etiology and disease outcomes among this unique group of patients remain poorly understood.^18,20,21 Herein, we present the largest multicenter study to date comparing demographic and survival outcomes in young versus old OCSCC patients. In this study, young patients aged ≤ 40 years represented 4.55% of our total cohort. This proportion is similar to that reported in literature.^17
-20,22 This age cutoff was selected both to describe a patient population less exposed to the traditional risk factors of smoking and alcohol consumption as well as to align with similar studies in the literature.

With regard to cancer subsite, oral tongue was the most common in our study. This aligns with other studies in the literature. ^12,18,23 There has been some conflicting evidence within the literature as to whether the oral tongue subsite has an improved or worsened survival when compared with other subsites.^24
-27 In our analysis, the floor of mouth, buccal, and retromolar trigone were found to have a worse prognosis when compared with oral tongue.

A number of tumor risk factors were found in similar proportions among young and old OCSCC patients, including: lymphovascular invasion, tumor thickness, post-operative margin status, nodal stage, and extranodal extension (ENE). Our findings mirror those reported in previous literature.^19,28 Histological differences were seen in tumor grade, bony invasion, and perineural invasion. Favorable tumor grade was noted to be more common among young compared to old OCSCC patients (35.5% well-differentiated among young vs 29.5% among old, P < 0.05). This is in contrast to previous studies, which found no difference in tumor grade between young and old patients.^28,29 As for bony invasion, young patients within this study had less when compared to their old counterparts (7.3% bony invasion at presentation vs 17.0%, P < 0.003). While the presence of bony invasion is reflected in a patient’s T stage, some studies have suggested that bony invasion (in particular, medullary bone invasion) is an independent risk factor for poor prognosis.³⁰ Interestingly, our findings are in contrast to those of Mneimneh et al, who found no difference in the proportion of bony invasion between young and old OCSCC patients.²⁷ Perineural invasion is a widely recognized indicator of poor prognosis, strongly correlating with aggressive tumor behavior, disease recurrence, and increased morbidity and mortality in OCSCC patients.^31,32 Perineural invasion was more common in young compared to old patients in our analysis (39.9% vs 32.6%, P < 0.05). The higher propensity for perineural invasion in young patients has been similarly reported in other recent literature.^18,19

In our cohort, young patients were more likely to present at earlier clinical stages, with a higher percentage classified as T1 versus T3-4 (44.0% in young vs 31.2% in old). The findings of previous literature examining this variable are mixed, with some smaller studies reporting higher stage at presentation for young patients, and others reporting similar stage disease.^28,33
-39 Both scenarios are plausible as young patients may be more proactive in seeking medical attention and thus present for diagnosis at an earlier stage, conversely, health professionals may have a higher threshold for suspicion in young, healthy patients and therefore be less likely to refer for a biopsy or further investigations.

When it comes to treatment, surgery is considered the standard of care for most tumors of the oral cavity and was used as the primary treatment modality in virtually all studies in the literature on this topic. Many studies have demonstrated improved local, regional, and distant disease outcomes when surgery is used as the primary modality of treatment. That said, patients must meet selected health parameters to be considered good candidates for surgery, particularly in the setting of extended anesthetic times for free tissue transfer. In our patient population, young OCSCC patients underwent primary surgery (with or without radiotherapy) more commonly compared to old patients (97.1% vs 88.3%). The proportion of patients receiving post-operative radiation therapy was not statistically different between young and old patients.

Chemoradiotherapy is used in the post-operative setting in the case of positive margins or ENE. It is well known to improve locoregional and distant metastatic disease in these settings.^40,41 That said, the use of chemotherapy dramatically increases treatment morbidity, and can even hasten mortality in older patients.^40,41 In our study, we found that young patients were more likely to receive adjuvant chemotherapy as part of their definitive treatment (excluding salvage/palliative treatment) than old patients (21.4% vs 12.8%), despite similar margin status and ENE status among the two cohorts. This may highlight the improved baseline health status of the younger patients, allowing them to be candidates for escalated treatment.

Our survival outcomes differed between univariate and multivariate analysis. Using univariable analyses, we found that young OCSSC patients had improved OS, with a higher proportion of young patients (82.0%) alive at the end of the study compared to old patients (70.6%, HR 1.75, CI 1.05-2.91). There were a number of risk factors that were favorable in young OCSCC patients when compared with older patients. These included lower tumor grade, earlier T stage at presentation, less frequent bony invasion, more curative intent treatment, more surgery as the primary modality of treatment, and more use of chemotherapy. Furthermore, one might expect old patients to die of other causes. Notably, we found that 29.0% of the old patients died from causes unrelated to OCSCC, compared to young patients (5.9%).

Our study, however, did not demonstrate a statistically significant difference in OS (Old patients HR 0.94, CI 0.61-1.45) or DFS (Old patients HR 1.02, CI 0.66-1.56) in multivariable analyses. This absence of a survival advantage for young patients despite advantages in tumor grade, T stage, ECOG status, smoking and alcohol consumption, CDCI, less frequent bony invasion, more curative intent treatment, more surgery as the primary modality of treatment, and more use of chemotherapy is noteworthy. There are a number of possible explanations for this finding. It could be the case that perineural invasion (the only disadvantageous tumor characteristic seen more frequently in young patients) has an outsized influence on survival. It may imply the existence of inherent factors specific to young patients, beyond the traditional risk factors of smoking and alcohol, which were not captured in our study, and which may contribute to poorer outcomes. It is also possible that despite our large sample size, our study was still underpowered to detect a true difference in survival between young and old patients. Additionally, we were not able to perform a subgroup analysis of young, non-smoking patients due to sample size constraints, and thus, the lack of difference in OS and DFS between young and old groups may be driven by poorer outcomes among young smokers compared to young non-smokers. Our findings of similar survival in young and old OCSCC patients mirror those of selected previous studies^{21,23,25,35,42
-44} but also contradict three studies that demonstrated worsened survival^18,27,45 and four studies that demonstrated improved survival in the young OCSCC patient population.^12,17,20,28 That said, these studies with findings that contradicted ours were either small, single institution studies or from large national databases with data less granular than our study.

Further investigation into the differences in tumor biology between young and old patients could help shed light on the discrepancy in outcomes between these studies. Several distinct driver genes have been suggested to be responsible for promoting the development of OCSCC, including TP53, CDKN2A, CASP8, NOTCH1, FAT1, ATXN1, and CDC42EP1 (33146897). Some studies demonstrate no difference in gene mutations between young and old patients (33210069, 24874835). A recent study suggests that OCSCC of the oral tongue in young patients has a significantly smaller somatic mutational burden as compared to that of old patients, even when correcting for differences in tobacco use (33146897). One study had suggested a potential viral role to account for the smaller number of mutations, but no viral etiology was identified (24954188). More research is warranted to elucidate any potential differences in genetic alterations driving carcinogenesis between young and old patients.

This population-based study is strengthened as it is the largest multi-center study providing a comprehensive overview of young OCSCC demographic, clinicopathologic, and survival outcomes. Within Canada’s universal health care system, oncological care is delivered at a regional level, with almost all patients being treated at tertiary centers by a multidisciplinary head and neck oncology team. This provides more consistency in our analysis as compared to that obtained from other jurisdictions and registries that may not have a similar structure in oncological care. While our multi-institutional study benefits from a robust cohort size, it is important to acknowledge certain limitations, including those that are inherent to the use of retrospective datasets. Despite implementing a careful standardization process for data collection, variations in treatment regimens across different institutions and provinces in Canada may introduce bias. Furthermore, our study is limited to patients treated at Canadian centers, which may restrict the generalizability of our findings. Finally, certain subgroup analyses that were desired were not performed due to sample size limitations. These include young non-smokers versus other, female young non-smokers and analyses between different oral cavity subsites. Although this approach allowed us to maintain a robust sample size, it precludes the identification of potential differences that may influence our results. A potential bias to consider is that nearly 30% of the older cohort dies from other causes, which may affect recurrence rates. It is possible that a substantial proportion of these patients, had they lived longer, might have experienced recurrence, thereby influencing the observed equivalence in recurrence rates between younger and older patients despite differences in initial disease stage.

Conclusion

In conclusion, this study demonstrates that young age is not a significant hazard for worse OS or DFS in patients with OCSCC. Despite presenting with more favorable clinical staging, risk factor profiles, and tumor pathological characteristics, young patients do not exhibit a better prognosis compared to older patients with OSCC after adjustment in multivariable models. Future investigation should focus on conducting subgroup analyses among OSCC subsites and explore potential differences in tumor biology between young and old patients.

Supplemental Material

sj-docx-1-ohn-10.1177_19160216251351562 – Supplemental material for Oral Cavity Squamous Cell Carcinoma in Young Patients: A Multi-Institutional Study of the Canadian Head & Neck Collaborative Research Initiative

Supplemental material, sj-docx-1-ohn-10.1177_19160216251351562 for Oral Cavity Squamous Cell Carcinoma in Young Patients: A Multi-Institutional Study of the Canadian Head & Neck Collaborative Research Initiative by Xinyuan Hong, Alexandra E. Quimby, Dorsa Mavedatnia, A. Travis Pickett, Martin Corsten, Tinghua Zhang, Angelina Tohme, Stephanie Johnson-Obaseki, Carlos Khalil, Mark Khoury, Antoine Eskander, Hesameddin Noroozi, David Goldstein, John De Almeida, James Fowler, S. Danielle MacNeil, Anthony C. Nichols, Joseph Dort, Robert Hart, Wayne Matthews, Apostolos Christopoulos, Gabriel Dayan, Houda Bahig, Michael P. Hier, Khalil Sultanem, Gregoire B. Morand, Brigitte Routhier-Chevrier, Zhi Shuo Zhang, Mathieu Belzile, Hamza Laref, Marie-Jo Olivier and Anastasios Maniakas in Journal of Otolaryngology - Head & Neck Surgery

Footnotes

Acknowledgements

Each author has played a significant role in the conception, design, execution, or interpretation of the work and has approved the final version of the manuscript for publication. All contributing authors have been listed.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval and Informed Consent

Ethical approval for this study was obtained from the Clinical Trials of Ontario (CTO) and a waiver for informed consent was granted.

ORCID iDs

Angelina Tohme

Gabriel Dayan

Supplemental Material

Supplemental material for this article is available online.

References

Siegel

Miller

Jemal

Cancer statistics. 2017. CA: Cancer J Clin 2017;67(1):7-30.

Canadian Cancer Statistics Advisory Committee. Canadian Cancer Statistics 2018. Canadian Cancer Society; 2018. Accessed January 23, 2019, cancer.ca/Canadian-Cancer-Statistics-2018-EN.

Blot

McLaughlin

Winn

, et al. Smoking and drinking in relation to oral and pharyngeal cancer. Cancer Res. 1988;48(11):3282–3287.

Peters

McClean

Marsit

, et al. Glutathione S-transferase polymorphisms and the synergy of alcohol and tobacco in oral, pharyngeal, and laryngeal carcinoma. Cancer Epidemiol Biomarker Prev. 2006;15(11):2196–2202.

Kraunz

McClean

Nelson

, et al. Duration but not Intensity of alcohol and tobacco exposure predicts p16-INK4A homozygous deletion in head and neck squamous cell carcinoma. Cancer Res. 2006;66(8):4512-4515.

Lubin

Purdue

Kelsey

, et al. Total exposure and exposure rate effects for alcohol and smoking and risk of head and neck cancer: a pooled analysis of case-control studies. Am J Epidemiol 2009;170(8):937-947.

Canadian Cancer Society. Risk factors for oral cancer. Canadian Cancer Society; 2019. Accessed January 23, 2019. http://www.cancer.ca/en/cancer-information/cancer-type/oral/risks

Warnakulasuriya

Global epidemiology of oral and oropharyngeal cancer. Oral Oncol. 2009;45(4-5):309-316.

Satgunaseelan

Allanson

Asher

, et al. The incidence of squamous cell carcinoma of the oral tongue is rising in young non-smoking women: An international multi-institutional analysis. Oral Oncol. 2020;110 :104875.

10.

Toporcov

Znaor

Zhang

, et al. Risk factors for head and neck cancer in young adults: a pooled analysis in the INHANCE consortium. Int J Epidemiol. 2015;44(1):169-185.

11.

Harris

Kimple

Hayes

, et al. Never-smokers, never-drinkers: unique clinical subgroup of young patients with head and neck squamous cell cancers. Head Neck. 2010;32(4):499-503.

12.

Fan

Zheng

Mao

, et al. Survival analysis of oral squamous cell carcinoma in a subgroup of young patients. Asian Pac J Canc Prev. 2014;15(20):8887-8891.

13.

Wey

Lotz

Triedman

LJ.

Oral cancer in women nonusers of tobacco and alcohol. Cancer. 1987;60(7):1644-1650.

14.

Yesensky

Hasina

Wroblewski

, et al. Role of dental hardware in oral cavity squamous cell carcinoma in the low-risk nonsmoker nondrinker population. Head Neck. 2018;40(4):784-792.

15.

Perry

Zammit

Lewandowski

, et al. Sites of origin of oral cavity cancer in nonsmokers vs smokers: possible evidence of dental trauma carcinogenesis and its importance compared with human papillomavirus. JAMA Otolaryngol Head Neck Surg. 2015;141(1):5-11.

16.

Scheidt

Yurgel

Romanini

, et al. Oral squamous cell carcinoma from users and nonusers of tobacco and alcohol: clinicopathologic features and immunoreactivity of VEGF, caspase-3, and p53. Appl Immunohistochem Mol Morphol. 2013;21(2):148-153.

17.

Mukdad

Heineman

Alonso

Badran

Kuan

St. John

MA.

Oral tongue squamous cell carcinoma survival as stratified by age and sex: A surveillance, epidemiology, and end results analysis. Laryngoscope. 2019. doi:10.1002/lary.27720

18.

Valero

Yuan

Zanoni

, et al. Young non-smokers with oral cancer: what are we missing and why? Oral Oncol. 2022;127:105803.

19.

Lee

Ramirez

Lee

, et al. Survival of young vs old patients with oral cavity squamous cell carcinoma: a meta-analysis. Laryngoscope. 2021;131(6), 1310–1319. doi:10.1002/lary.29260

20.

Annertz

Anderson

Biörklund

, et al. Incidence and survival of squamous cell carcinoma of the tongue in Scandinavia, with special reference to young adults. Int J Cancer. 2002;101:95-99.

21.

Patel

Carpenter

Tyree

, et al. Incidence increasing of oral tongue squamous cell carcinoma in young white women, age 18-44 years. J Clin Onc. 2011;29(11):1488-1494.

22.

Wang

, et al. The impact of age on oral squamous cell carcinoma: A longitudinal cohort study of 2,782 patients. Oral Dis. 2019;25(3):730–41. doi: 10.1111/odi.13015

23.

Park

Sun

Cho

, et al. Clinical outcome of squamous cell carcinoma of the tongue in young patients: a stage-matched comparative analysis. Clin Exp Otorhinolar. 2010;3(3):161-165.

24.

Byers

RM.

Squamous cell carcinoma of the oral tongue in patients less than thirty years of age. Am J Surg. 1975;130:475-478.

25.

Sun

Fang

Guo

A comparison of oral squamous cell carcinoma between young and old patients in a single medical center in China. Int J Clin Exp Med. 2015;8:12418-12423.

26.

McGregor

Davis

Robins

RE.

Squamous cell carcinoma of the tongue and lower oral cavity in patients under 40 years of age. Am J Surg. 1983;146:88–92

27.

Mneimneh

Ghossein

, et al. Clinicopathologic characteristics of young patients with oral squamous cell carcinoma. Head Neck Pathol. 2021;15, 1099-1108. doi:10.1007/s12105-021-01320-w

28.

Oliver

Chang

, et al. Survival of oral tongue squamous cell carcinoma in young adults. Head Neck. 2019. doi:10.1002/hed.25772

29.

Ansarin

De Berardinis

Corso

, et al. Survival outcomes in oral tongue cancer: a mono-institutional experience focusing on age. Front Oncol. 2021;11:616653. doi: 10.3389/fonc.2021.616653

30.

Ebrahimi

Murali

Gao

Elliott

Clark

JR.

The prognostic and staging implications of bone invasion in oral squamous cell carcinoma. Cancer. 2011;117(19):4460-4467.

31.

Woolgar

JA.

Histopathological prognosticators in oral and oropharyngeal squamous cell carcinoma. Oral Oncol. 2006;42(3):229-239.

32.

Fielder

Off

Eichberger

, et al. OSCC in never-smokers and never-drinkers is associated with increased expression of tumor-infiltrating lymphocytes and better survival. Cancers. 2023;(15):2688-2703.

33.

Funk

Karnell

Robinson

Zhen

Trask

Hoffman

HT.

Presentation, treatment, and outcome of oral cavity cancer: a National Cancer Data Base report. Head Neck. 2002;24(2):165-180.

34.

Garavello

Spreafico

Gaini

RM.

Oral tongue cancer in young patients: a matched analysis. Oral Oncol. 2007;43(9):894- 897.

35.

Kourelis

Tsue

Girod

Tawfik

Sykes

Shnayder

Negative prognostic factors for head and neck cancer in the young. Surgery. 2013;18(2):459-464.

36.

Venables

Craft

IL.

Carcinoma of the tongue in early adult life. Br J Cancer. 1967;21(4):645-650.

37.

Iype

Pandey

Mathew

Thomas

Sebastian

Nair

MK.

Oral cancer among patients under the age of 35 years. J Postgrad Med. 2001;47(3):171-176.

38.

Pitman

Johnson

Wagner

Myers

EN.

Cancer of the tongue in patients less than forty. Head Neck. 2000;22(3):297-302.

39.

Veness

Morgan

Sathiyaseelan

Gebski

Anterior tongue cancer: age is not a predictor of outcome and should not alter treatment. ANZ J Surg. 2003;73(11):899-904.

40.

Bernier

Domenge

Ozsahin

, et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med. 2004;350:1945-1919

41.

Cooper

Pajak

Forastiere

, et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med. 2004;350:1937-1944.

42.

Choi

Song

Choi

, et al. Comparison of squamous cell carcinoma of the tongue between young and old patients. J Pathol Transl Med. 2019;53(6):369-377. doi: 10.4132/jptm.2019.09.16

43.

Kwon

Lee

Choi

, et al. Clinicopathological characteristics of young never smokers females with oral cavity squamous cell carcinoma: a STROBE compliant retrospective observational study. Med Open. 2021;100(5):1-7.

44.

Zhang

Wang

, et al. Clinicopathological characteristics and outcomes of squamous cell carcinoma of the tongue in different age groups. Head Neck. 2017;39(11):2276-2282.

45.

Sarkaria

Harari

PM.

Oral tongue cancer in young adults less than 40 years of age: rationale for aggressive therapy. Head Neck. 1994;16(2):107-111.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB