Incidence and survival in sinonasal carcinoma: a Danish population-based, nationwide study from 1980 to 2014

Abstract

Background

Sinonasal cancers are rare and comprise <1% of all malignancies. This study describes incidence and survival in sinonasal carcinomas in Denmark from 1980 to 2014.

Methods

All patients registered in the Danish Cancer Registry in the period were included. Age-adjusted incidence rate, average annual percentage change, and relative survival were calculated. Age-period-cohort models were constructed.

Results

1,720 patients with sinonasal carcinoma (median age 67 years, 63% males) were identified. There was no significant change in age-adjusted incidence; 0.70 in 1980 to 0.43 per 100,000 in 2014 (p > .05). Relative 5- and 10-year survival were 52% and 40% for men, 58% and 42% for women. An increase in 5-year survival from 1980 to 2014 from 46% to 65% (p < .05) was found. Nasal carcinomas had a significantly better relative survival compared to sinus carcinoma, as did squamous cell carcinomas when compared to neuroendocrine malignancies.

Conclusion

In Denmark between 1980 and 2014, the incidence of sinonasal carcinomas has been stable and the relative survival has increased significantly.

Introduction

Cancers of the nasal cavity and paranasal sinuses are rare and comprise 5% of head and neck cancers and less than 1% of all malignant tumors [1,2]. Sinonasal cancers consist of many different histological types dominated by cancers of epithelial histology [1,2]. Squamous cell carcinomas (SCC) are the most common, followed by adenocarcinomas (AC) [1,3,4]. Exposure to several substances are well-known risk factors for developing sinonasal cancer, such as dust derived from wood (AC), leather (mainly AC), nickel and chromium and tobacco smoking (SCC) [5,6]. In recent years a new risk factor has emerged as the human papillomavirus (HPV), which is seen in a subset of sinonasal SCC with a probably more favorable prognosis than HPV-negative SCC [7,8]. Infection with HPV is a well-established causative agent in the rise in incidence of certain other head and neck cancers such as the oropharyngeal SSC, having a more favorable five-year overall survival [9,10].

Due to vague and unspecific symptoms sinonasal tumors are often diagnosed at an advanced stage [1,11], and therefore rarely amenable to a complete surgical resection, reflected in an overall poor prognosis [2,12]. Sinonasal tumors are located in an area of complex anatomy with a close relation to important structures such as the eyes and brain, which poses a challenge in the planning of surgery as well as radiotherapy [1,13].

The current treatment for sinonasal carcinomas is surgery for low-stage cancers (I/II) and multimodal therapy (surgery ± radiotherapy) for cancers in advanced stages (III/IV) [1,2,12,14]. The current Danish national guidelines recommends surgery as the primary treatment of sinonasal carcinoma, with the addition of postoperative radiation therapy in case of positive margins, stage III–IV tumors and adenoid cystic carcinoma. Radiation therapy can be used as a primary treatment modality should the patient request this or if surgery is not possible due to comorbidities. Chemotherapy is only recommended in the treatment of undifferentiated carcinomas [14].

Studies examining incidence and survival in sinonasal carcinoma for histological types as well as anatomical locations are limited. The aim of this study was to describe a putative changing trend in sinonasal carcinoma in a large nationwide cohort from Denmark from 1980 to 2014, and to evaluate changes in survival based on histology, location, and gender.

Material and methods

This study was conducted from a source population from the Danish Cancer Registry (DCR), which contains data on cancers diagnosed in Denmark since 1943, to which reporting was mandatory from 1987 [15]. Data on age at diagnosis were provided from the DCR, vital status and dates of emigration were obtained from the Central Population Register (CPR), which is linked to the DCR [15,16]. The CPR issues all Danish residents with a unique personal identification number (the CPR number) since its establishment in 1968. This CPR number enables accurate linkage of data from one or more registers with register-based information at an individual level and provides unique possibilities for nationwide epidemiological studies [16].

Age-specific population counts were derived from the National Statistical Database.

During the study period, two different versions of the International Classification of Diseases (ICD) were used; until 2004 patients were coded according to ICD-O-1 (International Classification of Diseases for Oncology). These codes were converted to ICD-10 codes via ICD-O-2 and ICD-O-3 [15]. All patients registered with sinonasal carcinomas in Denmark from 1980 to 2014 were included. Based on the ICD-10 codes, patients were divided into five locations: Maxillary sinus (DC310), ethmoid and frontal sinus (DC311 and DC312), sphenoid sinus (DC313), unspecified sinus (DC318 and DC319), and nasal cavity (DC30, DC300, DC300A, DC300B, and DC300C). The histological/morphological information on tumors were from the ICD-O-3 classification, which is also a part of the DCR [15].

The histological information on the carcinomas was grouped into seven groups; SCC, non-salivary type adenocarcinoma (NSG-AC), salivary gland type adenocarcinoma (SG-AC), neuroendocrine carcinoma (NEC), esthesioneuroblastoma (ENB), unspecified carcinoma (UC) and other.

The study was approved by The Danish Data Protection Agency.

Statistical analyses

Statistical analyses were performed in R statistics version 3.3.3. The average annual percentage change (AAPC) in incidence was calculated using the Joinpoint Trend Analysis Software v. 4.2.0.2 (National Cancer Institute) with growth assumed to be logarithmic and following the formula ln(y) = xb and allowing a maximum of five joinpoints in the analysis of AAPC from 1980 to 2014. Five-year age-specific crude incidence rates and age adjusted incidence rates per 100,000 were calculated using the direct method with R Statistics and the package EpiTools using the WHO World Standard Population as reference [17].

The relative survival rate was calculated as the ratio of the observed survival rate to the expected survival rate in Denmark, matched by age, sex and calendar year with the R package relsurv [18]. Relative survival was defined as the all-cause observed survival in the cancer population under study divided by the expected survival of a comparable group in the general population. Expected survival was estimated using the Ederer II method [19]. Patients alive at the last date of follow-up were censored at this date. We tested the proportional hazards assumption of the relative survival models by Brownian Bridge statistics [20].

Effects of age, calendar period and birth cohort on incidence was evaluated using age-period-cohort (APC) models [21]. We restricted the APC analyses to persons aged 30–84 to avoid statistical instability due to the small numbers of cancer cases among those younger than 30 and those older than 84. The study populations were categorized into five-year age groups (from 30 to 84 years) and five -year calendar period groups (1980–1984, 1985-1989, 1990–1994, 1995–1999, 2000–2004, 2005–2009, 2010–2014) according to the date of diagnosis.

The R package Epi was used to construct APC models with a two-sided probability [21]. The reference year for cohort effects was 1900 and 1982.5 for period effects.

We considered a p value < .05 significant.

Results

A total of 1720 patients with sinonasal carcinoma were included. The median age at diagnosis was 67 years (range 2–95) and the majority of patients were males (62%) (Table 1 and Figure 1(A)). The most frequent location for sinonasal carcinoma was the nasal cavity comprising 71% of all tumors followed by the maxillary sinus (17%) (Table 1 and Figure 1(B)). The most common histological type was SCC (61%) with NSG-AC (22%), and UC (5%) being the second and third most common, respectively (Table 1 and Figure 1(C)).

Figure 1. Pie charts illustrating the fraction of (A) men and women, (B) locations, and (C) histological types. (D) Total age-adjusted incidence rates per 100,000 for the total population for each year. (E) Median age at diagnosis for the different years. SCC: Squamous cell carcinoma; AC: adenocarcinoma; SG-AC: salivary gland type adenocarcinoma; NEC: neuroendocrine carcinoma; ENB: esthesioneuroblastoma.

Table 1. Characteristics of patients with sinonasal carcinoma in Denmark and changes in incidence and survival.

					UNIVARIATE		MULTIVARIATE
	AAPC (95% CI)	p	Cases n (%)	Events (deaths)	HR (95% CI)	p	HR (95% CI)	p
Male	−0.3 (−0.9;0.3)	.3	1078 (63)	857	Ref.	–	Ref.	–
Female	1.1 (−0.5;2.8)	.2	642 (37)	499	1.06 (0.94–1.18)	.34	1.08 (0.97–1.21)	.17
SCC	0.0 (−0.7;0.7)	.9	1046 (61)	823	Ref.	–	Ref.	–
NSG-AC	−2.8 (−5.9;0.5)	.1	374 (22)	296	0.91 (0.79–1.04)	.15	0.89 (0.78–1.02)	.10
UC	0.8 (−5.3-7.2)	.8	94 (5)	78	1.84 (1.45–2.32)	.00	1.84 (1.45–2.34)	.00
SG-AC	4.6 (−3.4:13.1)	.3	88 (5)	70	1.12 (0.88–1.43)	.36	0.9 (0.7–1.16)	.41
NEC	9.4 (0.2;19.4)	.0	40 (2)	34	1.84 (1.31–2.6)	.00	1.62 (1.14–2.28)	.01
Other	−1.0 (−7.8;6.3)	.8	11 (0.6)	9	1.01 (0.53–1.95)	.97	0.99 (0.51–1.92)	.97
ENB	3.7 (0.2;19.4)		67 (4)	46	1.7 (1.26–2.29)	.00	1.46 (1.08–1.99)	.01
DC300	−0.3 (−0.2;1.4)	.7	1217 (71)	913	Ref.	–	Ref.	–
DC310	−2.9 (−4.4; −1.5)	.0	298 (17)	272	1.95 (1.7–2.23)	.00	2.02 (1.76–2.33)	.00
DC311 + 12	−5.7 (−13.5:2.8)	.2	38 (2)	34	1.59 (1.13–2.24)	.01	1.63 (1.15–2.3)	.01
DC313	4.7 (−0.6;10.3)	.1	14 (0.8)	13	2.2 (1.27–3.8)	.00	2.41 (1.38–4.22)	.00
DC318 + 19	4.5 (0.2;8.9)	.0	153 (9)	124	1.72 (1.43–2.08)	.00	1.78 (1.47–2.17)	.00
Total	–	–	1720 (100)	1356	Ref.	–	Ref.	–
1980–1984	–	–	203 (12)	198	Ref.	–	Ref.	–
1985–1989	–	–	213 (12)	205	0.93 (0.76–1.13)	.47	0.95 (0.78–1.16)	.64
1990–1994	–	–	229 (13)	213	0.94 (0.77–1.14)	.51	1 (0.82-1.21)	.97
1995–1999	–	–	265 (15)	225	0.76 (0.62–0.92)	.00	0.81 (0.67–0.98)	.03
2000–2004	–	–	245 (14)	200	0.92 (0.75–1.12)	0.38	1.06 (0.86–1.29)	.60
2005–2009	–	–	290 (17)	195	0.81 (0.66–0.99)	.04	0.94 (0.77–1.15)	.54
2010–2014	–	–	259 (15)	106	0.62 (0.49–0.79)	.00	0.71 (0.56–0.91)	.01

AAPC; average annual percentage change in age-standardized incidence rate. HR; hazard ratio. CI; confidence intervals. SCC; squamous cell carcinoma. NSG-AC; non-salivary gland adenocarcinoma. SG-AC; salivary gland type adenocarcinoma. NEC; neuroendocrine cancers. ENB; esthesioneuroblastomas. UC; carcinoma UNS. DC300; nasal cavity. DC310; maxillary sinus. DC311 + 312; ethmoid and frontal sinus. DC313; sphenoid sinus. DC318 + 319; unspecified sinus.

Bold indicates significant values.

Incidence trends of sinonasal cancer in Denmark from 1980 to 2014

The number of cases and the incidence of sinonasal cancer were increasing with increasing age (Figure 1(D)). In general, the incidence of sinonasal carcinoma was close to zero before the age of 30. The median age did not change throughout the period (Figure 1(E)).

The total age-adjusted incidence rate (AAIR) for sinonasal cancers was 0.70 per 100,000 in 1980 and 0.47 per 100,000 in 2014 (p = .6), corresponding to 50 cases in 1980 and 52 cases in 2014.

There was no significant change in total AAPC which was 0.2 (95% CI −0.5;0.8), nor in the AAPC for men or women; −0.2 (95% CI −0.8;0.4) and 1.0(95% CI −0.5;2.6), respectively. AAPC was 2.3 (95% CI −2.0;6.8) for the period 1987–2014 when reporting to the DCR was mandatory. The AAIR of sinonasal carcinoma was 0.4 per 100,000 (95% CI 0.3;0.6) in 2014, with no significant difference between men (0.6 per 100,000 (95% CI 0.4;0.9)) and women (0.3 per 100,000 (95% CI 0.2;0.9)).

The AAPC of histological types was constant except for NEC where an increase in AAPC of 3.7 (95% CI 0.2;19.4) was observed (Table 1). The incidence of the most common histological type, SCC was 0.4 per 100,000 (95% CI 0.2;0,5) in 2014.

AAIR varied depending on the location; the incidence of nasal cavity tumors was 0.3 per 100,000 (95% CI 0.2;0.5) in 2014, which was significantly higher than the incidence of the second most frequent location, the maxillary sinus (AAIR 0.1 per 100,000 (95% CI 0.0;0.2)).

We observed a significant increase in incidence for nasal cavity tumors (AAPC 0.7 (95% CI 0.0;1.4)) and tumors in unspecified location (DC318 and DC 319) (AAPC 4.5 (95% CI 0.2;8.9)) and a decreased incidence for maxillary tumors (AAPC −2.9 (95% CI −4.4;-1.4)) (Table 1). There was no significant change in AAPC for tumors in the ethmoid/frontal sinus or sphenoid sinus.

Survival in sinonasal cancers in Denmark from 1980 to 2014

The median survival in this cohort was 3.6 years (range: 0–38 years). For males, the five- and 10-year relative survival was 52% and 40%. For females the corresponding survival was 58% and 42%. There was no difference in survival when comparing men and women (Table 1 and Figure 2(C)).

Figure 2. Survival over time and number at risk. (A) Survival for different histologies. SCC: Squamous cell carcinoma; AC: adenocarcinoma; SG-AC: salivary gland type adenocarcinoma; NEC: neuroendocrine carcinoma; ENB: esthesioneuroblastoma. (B) Survival for different locations. DC300; nasal cavity, DC310; maxillary sinus, DC311/312; ethmoid and frontal sinus, DC313; sphenoid sinus, DC318/319; unspecified location. (C) Survival for men and women over time. (D) Survival according to year of diagnosis (in five year intervals).

The relative survival for nasal tumors was higher compared to cancers in the other locations, with a five- and 10-year survival of 63% (95% CI 60;66) and 49% (95% CI 45;53) respectively when compared to cancer of the maxillary sinus with a five- and 10-year survival of 30% (95% CI 24;36) and 20% (95% CI 15;26), respectively (Table 1 and Figure 2(B)).

The best relative survival between histological groups was seen in the SCC group with five year survival of 56% (95% CI 52;59). This was only significantly higher than the survival in the UC, which had a five year survival of 29% (95% CI 20;40) (Table 1 and Figure 2(A)). The latter containing the sinonasal undifferentiated carcinoma, which is known to carry a poor prognosis [22].

Five-year survival rate was significantly higher for 2010–2014 compared with 1980–1984; 65% (95% CI 58;73) and 46% (95% CI 39;54), respectively (Table 1 and Figure 2(D)).

A multivariate analysis showed no difference in survival for men and women. The prognosis was significantly better for SCC than UC, NEC and ENB (Table 1).

Age period cohort effects

When comparing men and women, there was no significant change in risk of developing sinonasal cancer in younger birth cohorts compared to birth cohorts from 1900 (Figure 3(B)). There were no significant period-effects (Figure 3(C)).

Figure 3. Total period effect. (A) Incidence for five-year age groups showing an increase in incidence in the older age groups. (B) Cohort effect and relative risk showing no significant difference in RR for the different cohorts. (C) Period and relative risk showing no significant difference in RR for the different periods. RR: relative risk; LCI: Lower confidens interval; UCI: Upper confidens interval.

Discussion

This comprehensive population-based study is the first to address the incidence and concurrent survival trends of sinonasal cancer in Denmark in the period 1980–2014. The most common histology was SCC, in similarity with other reports [5,23–25].

In consistency with previous studies [26], this study has demonstrated an overall unchanged incidence of sinonasal carcinomas from 1980 to 2014 and a majority of male patients [5,22–25], the latter perhaps attributable to the etiological association with occupational exposure to wood and leather dust particles in male dominated trades. Since reporting to the DCR became mandatory in 1987 AAPC was calculated for the period 1987–2014 as well, but no significant change was seen. As reporting to the Danish Cancer Registry became mandatory in 1987, this may explain the low AAIR in the first years of this study (Figure 1(D)).

The reported AAIR in this study is in line with previous studies demonstrating an incidence of 0.4–1.0 per 100,000 [1,5], as are the survival rates [22,26] and the distribution of cancers seen in this study; the most common location being the nasal cavity [22–24]. The increase in incidence of nasal cavity carcinoma may be due to earlier diagnosis with more localized disease with the introduction and increasing use of advanced imaging modalities from the 1980’s and the use of endoscopy for cancer screening [27,28]. The increase in NEC most likely reflects improvements in classification of these tumors including the introduction of immunohistechemisty in the histopathological examination.

This study demonstrates a statistically significant improvement in survival, perhaps due to better access to and refinement of imaging modalities as well as advances in surgical treatment. An improvement in survival from 1960s to 1990s has been demonstrated earlier [24]. More recent studies have not been able to find a significant improvement in survival [13,22,24,26]. The increase in five-year survival in this study seen in the period 2010–2014 may be explained by several factors. The expansion of the number of linear accelerators in Denmark from 2000 and onwards lead to shortening of the time from diagnosis to radiation therapy. This trend was further accentuated by the introduction of fast track programs for cancer treatment in 2007, speeding up the process of hospital referral, biopsy, imaging and treatment [29]. Improvements in imaging (PET/CT- and MRI-scans for staging and treatment planning) and in advanced radiation therapy planning and delivery (intensity modulated radiation therapy (IMRT), volumetric arc therapy (VMAT), and image guidance) have resulted in significant improvements in patient selection and radiation treatment precision. Additionally, in 2007, new national guidelines for sinonasal carcinoma recommended surgery as the treatment of choice for stage I and II disease, with radiotherapy as an adjuvant treatment in locally advanced disease [14]. An increase in survival was also seen in the period from 1995 to 1999, but no significant change was seen in the periods 2000–2004 or 2005–2009 in the multivariate analyses (Table 1)

Consistent with previous studies survival was related to location and tumors located in the nasal cavity had a better prognosis [22,24,30]. This may be due to earlier diagnosis of tumors in the nasal cavity owing to an earlier detection of symptoms such as unilateral nasal obstruction or epistaxis caused by nasal cavity tumors. Easier surgical access in the nasal cavity may also contribute to the better prognosis. D [5,13,22]. The better prognosis for SCC and adenocarcinoma in this study is also consistent with some previous studies [23]. A Finnish study of a comparable population, but with a smaller number of patients (n = 244) and including noncarcinoma cancers did not find a difference in disease-specific survival between histopathological types or different locations, perhaps attributal to the smaller number of patients. Survival was shown to be dependent on tumor stage [25].

The optimal treatment for malignant tumors of the nasal cavity and paranasal sinuses remains undetermined. Due to the low numbers of patients with sinonasal carcinoma, and the staggering amount of different histologies, prospective studies are scarce [2,3,12,22,26,29]. The diagnosis is most often made at an advanced stage [2,26]. The close proximity to important structures, e.g., eyes, optic nerves and brain, and the risk of local invasion also pose a challenge in planning both surgery and radiotherapy [13,23]. Image-guided endoscopic sinus surgery with the use of frozen specimens during surgery is now an accepted modality for treatment of malignant lesions of the sinonasal tract when feasible. This approach carries less operative morbidity compared with the traditional open surgery, which still has its place in the current treatment of sinonasal cancers [12,13,22,31]. Due to improved targeting, it has been suggested that intensity modulated radiation therapy (IMRT) may improve outcome compared to conventional radiotherapy [32].

The significant improvement in survival thus might be attributed to advances in treatment and the increasing role of surgeryas primary treatment as well as the introduction of a national accelerated cancer program in 2007. We cannot however conclude whether the improvement in survival could be attributed to earlier diagnosis due to the lack of information regarding tumor stage and treatment modality.

Our study has some limitations being of retrospective nature; the database does not hold complete information on treatment, TNM-stage or comorbidities. No pathology-specimens were reexamined due to anonymization of the data. The strengths of the study is that information was obtained from a national register, to which all reports of cancer became mandatory in 1987 allowing analysis of a large patient cohort, independent of individual or institutional biases as well as the structure of the Danish healthcare system, which is completely financed by taxes and in which all patients have equal access thus reflecting better the true incidence in the population.

In summary, sinonasal cancers are rare, consisting of many histological types of which the epithelial cancers are the most common. Sinonasal carcinomas are often diagnosed in an advanced stage and therefore associated with a poor prognosis.

Using population-based data from the DCR, we report an unchanged incidence of sinonasal cancer for men and women and a significant increase in survival in the period from 1980 to 2014. The study also demonstrates that the prognosis of this group of cancers is dependent on age, gender, histology, and location.

Disclosure statement

The authors report no conflicts of interest.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.