Key Points
Question
What is the survival associated with head and neck Merkel cell carcinoma according to American Joint Committee on Cancer 8th edition staging and treatment procedures in Canada?
Findings
This cohort study of 400 patients found that higher survival rates were associated with lower disease stage and multiple treatment modalities. Patients treated with surgery and radiotherapy had greater overall survival compared with those treated with surgery alone.
Meaning
These results suggest that disease stage and treatment modality are associated with survival, with multimodal treatment associated with greater overall survival across all stages.
Abstract
Importance
Merkel cell carcinoma (MCC) is an aggressive cutaneous neuroendocrine carcinoma. Due to its relatively low incidence and limited prospective trials, current recommendations are guided by historical single-institution retrospective studies.
Objective
To evaluate the overall survival (OS) of patients in Canada with head and neck MCC (HNMCC) according to American Joint Committee on Cancer 8th edition staging and treatment modalities.
Design, Setting, and Participants
A retrospective cohort study of 400 patients with a diagnosis of HNMCC between July 1, 2000, and June 31, 2018, was conducted using the Pan-Canadian Merkel Cell Cancer Collaborative, a multicenter national registry of patients with MCC. Statistical analyses were performed from January to December 2022.
Main Outcomes and Measures
The primary outcome was 5-year OS. Multivariable analysis using a Cox proportional hazards regression model was performed to identify factors associated with survival.
Results
Between 2000 and 2018, 400 patients (234 men [58.5%]; mean [SD] age at diagnosis, 78.4 [10.5] years) with malignant neoplasms found in the face, scalp, neck, ear, eyelid, or lip received a diagnosis of HNMCC. At diagnosis, 188 patients (47.0%) had stage I disease. The most common treatment overall was surgery followed by radiotherapy (161 [40.3%]), although radiotherapy alone was most common for stage IV disease (15 of 23 [52.2%]). Five-year OS was 49.8% (95% CI, 40.7%-58.2%), 39.8% (95% CI, 26.2%-53.1%), 36.2% (95% CI, 25.2%-47.4%), and 18.5% (95% CI, 3.9%-41.5%) for stage I, II, III, and IV disease, respectively, and was highest among patients treated with surgery and radiotherapy (49.9% [95% CI, 39.9%-59.1%]). On multivariable analysis, patients treated with surgery and radiotherapy had greater OS compared with those treated with surgery alone (hazard ratio [HR], 0.76 [95% CI, 0.46-1.25]); however, this was not statistically significant. In comparison, patients who received no treatment had significantly worse OS (HR, 1.93 [95% CI, 1.26-2.96)].
Conclusions and Relevance
In this cohort study of the largest Canada-wide evaluation of HNMCC survival outcomes, stage and treatment modality were associated with survival. Multimodal treatment was associated with greater OS across all disease stages.
This cohort study evaluates the overall survival of patients in Canada with head and neck Merkel cell carcinoma according to American Joint Committee on Cancer 8th edition staging and treatment modalities.
Introduction
Merkel cell carcinoma (MCC) is a cutaneous neuroendocrine carcinoma, thought to arise from Merkel cell mechanoreceptors in the basal layer of the epidermis.1,2,3 The incidence of MCC is low, reported at 0.7 cases per 100 000 people per year in the US; however, the number of cases has increased by 95% since 2000, a rate much higher than that of melanoma or other solid tumors.4 This increase is attributed largely to longer life spans, with other risk factors including UV light exposure and immunosuppression from malignant neoplasms, organ transplant, and HIV infection.1,2,4,5 In the northern hemisphere where there are lower levels of UV exposure, MCC cases are also associated with genetic alterations from the integration of the Merkel cell polyomavirus genome.1,2,5
Merkel cell carcinoma is an aggressive disease, with nearly one-third of patients presenting with nodal or distant metastatic disease.6 The cohort used to develop the most recent American Joint Committee on Cancer (AJCC) staging demonstrated 5-year overall survival (OS) estimates of 51%, 35%, and 14% for local disease, nodal involvement, and distant metastatic disease, respectively.6 Low incidence rates and the absence of prospective trials limit treatment recommendations based on retrospective studies, which are often from single institutions with records spanning several decades. Most larger studies are based on administrative data, which lack granular clinical data. Treatment options predominantly include surgery with wide local excision plus sentinel lymph node biopsy with or without nodal dissection, radiotherapy alone, or a combination of these modalities for locoregional disease, with chemotherapy and, more recently, immunotherapy as options for metastatic or unresectable locally advanced MCC.3
Given the predilection of MCC for sun-exposed areas of the body, nearly 50% of primary tumors are found in the head and neck region. Furthermore, MCCs of the head and neck region, particularly MCC of the lip, have been associated with greater mortality compared with other tumor sites.3,7 Poorer prognosis may be associated with close and frequently positive surgical margins from anatomical, functional, and cosmetic limitations as well as lack of correlational value of sentinel lymph node status in the head and neck compared with other sites due to more complex lymphatic anatomy.7,8,9,10 Given the differences in outcomes of head and neck MCC (HNMCC) compared with MCC in other locations, it is essential to separate out this population to appropriately guide clinical decision-making, give accurate prognostication to patients, and determine the effectiveness of various treatment modalities.
The Pan-Canadian Merkel Cell Carcinoma Collaborative is a large multicenter national registry of patients with MCC tumors and their treatment and survival data. This database is among few in the world to directly incorporate data from multiple institutions including detailed clinical data and the first Canada-wide data set to be reported, to our knowledge. Using this patient cohort, our primary objective was to evaluate survival outcomes of patients with HNMCC in Canada according to AJCC 8th edition (AJCC 8) staging and treatment modalities.
Methods
After ethics approval from the Ottawa Hospital research ethics board, a retrospective cohort study of patients with a diagnosis of MCC of the head and neck regions was conducted using the Pan-Canadian Merkel Cell Carcinoma Collaborative. A waiver to receiving informed patient consent had been granted by the ethics review board as all identifying parameters had been anonymized. This database contains retrospectively collected data on patients older than 18 years of age with a diagnosis of MCC between July 1, 2000, and June 30, 2018, from 10 university centers and 3 provincial cancer registries across Canada. Patients with other malignant neoplasms were excluded from the database. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
This cohort was derived from patients with the primary site of the malignant neoplasm in the face, scalp, neck, ear, eyelid, or lip. Patient demographics (age, sex, comorbidities, immunosuppression determined by record review, and diagnosis), tumor characteristics (location and AJCC 8 staging), treatments (surgery, radiotherapy, chemotherapy, and immunotherapy), and disease outcomes (disease recurrence, time of death, and cause of death) were collected.
The outcomes of interest were 5-year OS, 5-year disease-specific survival (DSS), and 5-year recurrence-free survival (RFS). Overall survival was defined as the proportion of patients alive at 5 years after diagnosis, DSS was defined as the proportion of patients who did not die of MCC at 5 years after diagnosis, and RFS was defined as the proportion of patients who were alive and free of disease at 5 years after diagnosis.
Statistical Analysis
Statistical analyses were performed from January to December 2022 using SAS, version 9.3 (SAS Institute Inc). Data were summarized using frequency and percentages. Overall survival, DSS, and RFS hazard ratios (HRs) with 95% CIs were calculated using Cox proportional hazards regression models, which were constructed using clinically relevant and statistically significant variables to identify factors associated with survival. The Kaplan-Meier method was used to demonstrate survival outcomes by disease extent, AJCC 8 staging, and treatment type and to calculate 5-year survival estimates to account for censoring. Multivariable logistic regression models using clinically relevant and statistically significant variables were created to identify potential factors associated with survival. Variables that were statistically significant using 1-sided and 2-sided P < .05 and had complete data were included in the model. The Charlson comorbidity score was not included in the adjusted analysis because most patients had a Charlson comorbidity score higher than 4, limiting the clinical significance of the comparison.
Results
Patients
The database had a total of 1052 patients, of whom 400 (234 men [58.5%]; mean [SD] age at diagnosis, 78.4 [10.5] years) received a diagnosis of HNMCC between July 2000 and June 2018 (eFigure 1 and eTable 1 in Supplement 1). Most patients had several comorbidities, with a Charlson comorbidity score of 4 or higher (314 [78.5%]), and 41 patients (10.3%) had prior immunosuppression (eTable 1 in Supplement 1). Primary tumor sites were most commonly on the face (excluding eyelid and lip) (248 [62.0%]), followed by the neck and scalp (85 [21.3%]). At initial diagnosis, most tumors were classified as AJCC 8 stage I disease (188 [47.0%]).
Treatment
Overall, 161 patients (40.3%) were treated with surgery and radiotherapy, 107 (26.8%) with surgery alone, and 96 (24.0%) with radiotherapy alone (Table 1). Among patients with stage I disease, surgery alone and surgery with radiotherapy were equally common treatments, followed by radiotherapy and observation (65 of 188 [34.6%], 67 of 188 [35.6%], 35 of 188 [18.6%], and 20 of 188 [10.6%], respectively). Among patients with stage II disease, surgery alone, radiotherapy alone, and surgery with radiotherapy were all used in similar frequencies (23 of 82 [28.0%], 25 of 82 [30.5%], and 29 of 82 [35.4%], respectively), while chemotherapy alone was only used for 2 of 295 patients (0.7%) with stages I and III diseases. The most common choice of treatment by stage was surgery and radiotherapy for stage III disease (60 of 107 [56.1%]) and radiotherapy alone for stage IV disease (12 of 23 [52.2%]).
Table 1. Treatment Characteristics.
| Treatment | Patients, No./total No. (%) | ||||
|---|---|---|---|---|---|
| Total (N = 400) | Tumor stage | ||||
| Stage I (n = 188) | Stage II (n = 82) | Stage III (n = 107) | Stage IV (n = 23) | ||
| Surgery alone | 107/400 (26.8) | 65/188 (34.6) | 23/82 (28.0) | 16/107 (15.0) | 3/23 (13.0) |
| Radiotherapy alone | 96/400 (24.0) | 35/188 (18.6) | 25/82 (30.5) | 24/107 (22.4) | 12/23 (52.2) |
| Surgery plus radiotherapy | 161/400 (40.3) | 67/188 (35.6) | 29/82 (35.4) | 60/107 (56.1) | 5/23 (21.7) |
| No treatment | 34/400 (8.5) | 20/188 (10.6) | 5/82 (6.1) | 6/107 (5.6) | 3/23 (13.0) |
| Other | 2/400 (0.5) | 1/188 (0.5) | 0 | 1/107 (0.9) | 0 |
| Type of surgery (n = 270) | |||||
| WLE | 140/270 (51.9) | 94/133 (70.7) | 29/52 (55.8) | 12/77 (15.6) | 5/8 (62.5) |
| WLE plus SLNB | 60/270 (22.2) | 24/133 (18.1) | 12/52 (23.1) | 23/77 (29.9) | 1/8 (12.5) |
| WLE plus nodal dissection | 64/270 (23.7) | 12/133 (9.0) | 10/52 (19.2) | 40/77 (51.9) | 2/8 (25.0) |
| Nodal dissection alone | 2/270 (0.7) | 1/133 (0.8) | 0 | 1/77 (1.3) | 0 |
| Other | 4/270 (1.5) | 2/133 (1.5) | 1/52 (1.9) | 1/77 (1.3) | 0 |
| WLE margins, cm (n = 264) | |||||
| <1 | 3/130 (2.3) | 1/59 (1.7) | 0 | 2/42 (4.8) | 0 |
| 1 | 86/130 (66.2) | 40/59 (67.8) | 20/25 (80.0) | 24/42 (57.1) | 2/4 (50.0) |
| 2 | 27/130 (20.8) | 13/59 (22.0) | 4/25 (16.0) | 10/42 (23.8) | 0 |
| 3 | 7/130 (5.4) | 3/59 (5.1) | 0 | 2/42 (4.8) | 2/4 (50.0) |
| >3 | 7/130 (5.4) | 2/59 (3.4) | 1/25 (4.0) | 4/42 (9.5) | 0 |
| Unknown | 134 | ||||
| WLE margin status (n = 264) | |||||
| Positive margins | 58/203 (28.6) | 22/104 (21.2) | 18/41 (43.9) | 16/53 (30.2) | 2/5 (40.0) |
| Negative margins | 145/203 (71.4) | 82/104 (78.8) | 23/41 (56.1) | 37/53 (69.8) | 3/5 (60.0) |
| Unknown | 61 | ||||
| Management after positive margins on WLE (n = 58) | |||||
| Radiotherapy alone | 29/53 (54.7) | 11/21 (52.4) | 9/17 (52.9) | 8/13 (61.5) | 1/2 (50.0) |
| Reexcision alone | 9/53 (17.0) | 7/21 (33.3) | 1/17 (5.9) | 1/13 (7.7) | 0 |
| Radiotherapy plus reexcision | 8/53 (15.1) | 2/21 (9.5) | 3/17 (17.7) | 2/13 (15.4) | 1/2 (50.0) |
| Observation | 7/53 (13.2) | 1/21 (4.8) | 4/17 (23.5) | 2/13 (15.4) | 0 |
| Unknown | 5 | ||||
| Neoadjuvant radiotherapy (n = 264) | 1 (0.4) | 1/133 (0.8) | 0 | 0 | 0 |
| Adjuvant treatment (n = 270) | |||||
| Radiotherapy | 152 (56.3) | 66/133 (49.6) | 28/52 (53.9) | 55/77 (71.4) | 3/8 (37.5) |
| Chemotherapy | 2 (0.7) | 1/133 (0.8) | 0 | 1/77 (1.3) | 0 |
| Chemotherapy plus radiotherapy | 9 (3.3) | 1/133 (0.8) | 1/52 (1.9) | 5/77 (6.5) | 2/8 (25.0) |
| No treatment | 107 (39.6) | 65/133 (48.9) | 23/52 (44.2) | 16/77 (20.8) | 3/8 (37.5) |
Abbreviations: SLNB, sentinel lymph node biopsy; WLE, wide local excision.
A 1-cm margin was most frequently chosen for wide local excision (86 of 130 [66.2%]) (Table 1). After surgical resection, 58 of 203 patients (28.6%) were found to have positive margins on pathologic examination. For all stages, the predominant management of positive surgical margins was radiotherapy (29 of 53 [54.7%]; stage I, 11 of 21 [52.4%]; stage II, 9 of 17 [52.9%]; stage III, 8 of 13 [61.5%]). In stage I disease, reexcision was also commonly performed (7 of 21 [33.3%]), while observation was more frequently used in higher stages (stage II, 4 of 17 [23.5%]; stage III, 2 of 13 [15.4%]). Adjuvant radiotherapy was administered to a large proportion of surgical patients in all stages (152 of 270 [56.3%]; stage I, 66 of 133 [49.6%]; stage II, 28 of 52 [53.9%]; stage III, 55 of 77 [71.4%]; stage IV, 3 of 8 [37.5%]).
Survival
Five-year cumulative OS, DSS, and RFS outcomes are described in Table 2, eFigure 3 in Supplement 1, and the Figure. For the overall cohort, 5-year OS was 42.0% (95% CI, 35.9%-47.9%), 5-year DSS was 75.9% (95% CI, 69.3%-81.3%), and 5-year RFS was 50.1% (95% CI, 43.5%-56.4%) (Table 2). Five-year OS was highest among patients treated with surgery and radiotherapy (49.9% [95% CI, 39.9%-59.1%]). For AJCC 8 stage I, II, III, and IV diseases, OS rates were 49.8% (95% CI, 40.7%-58.2%), 39.8% (95% CI, 26.2%-53.1%), 36.2% (95% CI, 25.2%-47.4%), and 18.5% (95% CI, 3.9%-41.5%), respectively; DSS rates were 85.7% (95% CI, 76.1%-91.7%), 76.7% (95% CI, 61.5%-86.5%), 62.2% (95% CI, 48.0%-73.6%), and 38.0% (95% CI, 7.7%-69.4%), respectively; RFS rates were 58.8% (95% CI, 49.2%-67.2%), 47.1% (95% CI, 32.6%-60.3%), 40.1% (95% CI, 27.9%-52.1%), and 17.6% (95% CI, 1.0%-51.9%), respectively (Figure and Table 2).
Table 2. Five-Year Cumulative Survival by Disease Characteristics and Treatment.
| Characteristic | Overall survival | Disease-specific survival | Recurrence-free survival | |||
|---|---|---|---|---|---|---|
| Deaths, No./total No. (%) | 5-y Survival, % (95% CI) | MCC deaths, No./total No. (%) | 5-y Survival, % (95% CI) | Recurrences, No./total No. (%) | 5-y Survival, % (95% CI) | |
| Total | 210/400 (52.5) | 42.0 (35.9-47.9) | 60/400 (15.0) | 75.9 (69.3-81.3) | 141/400 (35.3) | 50.1 (43.5-56.4) |
| T stage | ||||||
| T1 | 93/189 (49.2) | 49.6 (40.5-58.1) | 15/189 (7.9) | 84.8 (75.1-91.0) | 57/189 (30.2) | 58.4 (48.8-66.9) |
| T2-3 | 31/69 (44.9) | 46.5 (30.5-61.0) | 9/69 (13.0) | 82.6 (66.6-91.4) | 24/69 (34.8) | 51.1 (35.1-65.1) |
| T4 | 9/11 (81.8) | 8.0 (0.1-39.3) | 2/11 (18.2) | 69.2 (22.7-91.3) | 4/11 (36.4) | 60.0 (25.3-82.7) |
| N stage | ||||||
| N0 | 136/275 (49.5) | 47.1 (39.5-54.2) | 28/275 (10.2) | 82.7 (75.0-88.2) | 89/275 (32.4) | 55.3 (47.4-62.6) |
| N1a | 12/30 (40.0) | 51.1 (29.7-69.0) | 7/30 (23.3) | 67.2 (42.4-83.2) | 10/30 (33.3) | 56.7 (33.7-74.4) |
| N1b | 48/74 (64.9) | 25.3 (13.9-38.3) | 20/74 (27.0) | 50.6 (30.6-67.6) | 34/74 (45.9) | 27.2 (13.5-42.8) |
| N2-3 | 6/12 (50.0) | 34.8 (5.5-68.0) | 2/12 (16.7) | 76.5 (34.1-93.5) | 3/12 (25.0) | 49.5 (9.9-80.7) |
| Tumor extent | ||||||
| Local | 132/266 (49.6) | 47.1 (39.5-54.4) | 25/266 (9.4) | 84.2 (76.5-89.6) | 85/266 (32.0) | 56.5 (48.5-63.8) |
| Regional | 59/105 (56.2) | 35.9 (24.7-47.3) | 26/105 (24.8) | 61.1 (46.6-72.7) | 44/105 (41.9) | 39.0 (26.8-51.1) |
| Metastatic | 16/23 (69.6) | 18.5 (3.9-41.5) | 7/23 (30.4) | 38.0 (7.7-69.4) | 9/23 (39.1) | 17.6 (1.0-51.9) |
| Tumor stage | ||||||
| Stage I | 93/188 (49.5) | 49.8 (40.7-58.2) | 14/188 (7.4) | 85.7 (76.1-91.7) | 57/188 (30.3) | 58.8 (49.2-67.2) |
| Stage II | 41/82 (50.0) | 39.8 (26.2-53.1) | 13/82 (15.9) | 76.7 (61.5-86.5) | 31/82 (37.8) | 47.1 (32.6-60.3) |
| Stage III | 60/107 (56.1) | 36.2 (25.2-47.4) | 26/107 (24.3) | 62.2 (48.0-73.6) | 44/107 (41.1) | 40.1 (27.9-52.1) |
| Stage IV | 16/23 (69.6) | 18.5 (3.9-41.5) | 7/23 (30.4) | 38.0 (7.7-69.4) | 9/23 (39.1) | 17.6 (1.0-51.9) |
| Treatment | ||||||
| Surgery alone | 56/107 (52.3) | 41.2 (29.3-52.6) | 15/107 (14.0) | 76.0 (60.7-86.0) | 41/107 (38.3) | 48.3 (35.6-59.8) |
| Radiotherapy alone | 55/96 (57.3) | 32.9 (21.5-44.7) | 16/96 (16.7) | 72.2 (56.6-83.0) | 26/96 (27.1) | 57.0 (42.1-69.4) |
| Surgery plus radiotherapy | 73/161 (45.3) | 49.9 (39.9-59.1) | 25/161 (15.5) | 77.3 (67.1-84.7) | 64/161 (39.8) | 49.9 (40.1-59.0) |
| No treatment | 25/34 (73.5) | 31.9 (15.6-49.5) | 3/34 (8.8) | 79.0 (45.5-93.2) | 9/34 (26.5) | 34.6 (9.3-62.2) |
Abbreviation: MCC, Merkel cell carcinoma.
Figure. Five-Year Survival Outcomes by American Joint Committee on Cancer 8th Edition Staging.
Univariable and multivariable analyses are described in Table 3 and Table 4. On univariable analysis, immunosuppression, T stage, N stage, and AJCC 8 stage were statistically significant for all survival outcomes (Table 3). Based on clinical significance, statistical significance, and completeness of data, variables included in the multivariable analysis were age, sex, immunosuppression, AJCC 8 stage, and treatment type. For all survival outcomes, increasing stage was associated with worse survival compared with stage I disease, although no statistically significant differences were observed in OS between patients with stage I and stage II disease (Table 4).
Table 3. Univariable Analysis of Survival Outcomes Adjusted by Charlson Score, Stage, and Treatment.
| Outcome | Overall survival | Disease-specific survival | Recurrence-free survival | |||
|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |
| Age | 1.04 (1.02-1.06) | <.001 | 0.99 (0.97-1.01) | .38 | 1.00 (0.99-1.01) | .78 |
| Sex | ||||||
| Male | 1.28 (0.93-1.76) | .13 | 2.16 (1.27-3.68) | .004 | 1.94 (1.47-2.56) | <.001 |
| Female | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| Charlson comorbidity score | ||||||
| 0-1 | 0.94 (0.61-1.44) | .77 | 2.50 (1.15-5.43) | .02 | 1.42 (0.85-2.40) | .18 |
| 2-3 | 0.30 (0.19-0.48) | <.001 | 0.68 (0.34-1.37) | .28 | 0.65 (0.40-1.04) | .07 |
| ≥4 | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| Immunosuppression | 1.66 (1.08-2.56) | .02 | 2.08 (1.29-3.35) | .003 | 2.28 (1.61-3.23) | <.001 |
| Primary location | ||||||
| Ear | 0.80 (0.52-1.22) | .30 | 0.69 (0.45-1.04) | .08 | 0.77 (0.44-1.34) | .36 |
| Eyelid | 0.44 (0.18-1.05) | .06 | NE | NE | 0.34 (0.06-1.86) | .21 |
| Lip | 0.65 (0.36-1.18) | .16 | 0.61 (0.17-2.18) | .45 | 0.69 (0.31-1.51) | .35 |
| Scalp or neck | 1.17 (0.89-1.54) | .27 | 1.21 (0.78-1.87) | .39 | 1.00 (0.69-1.45) | .99 |
| Face | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| T stage | ||||||
| T1 | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| T2 | 1.24 (0.85-1.82) | .26 | 2.37 (1.44-3.90) | <.001 | 1.65 (1.18-2.32) | .004 |
| T3 | 2.34 (1.35-4.08) | .003 | 3.39 (1.22-9.36) | .02 | 1.28 (0.49-3.36) | .62 |
| T4 | 3.08 (1.71-5.54) | <.001 | 3.59 (1.66-7.79) | .001 | 1.86 (0.96-3.58) | .07 |
| N stage | ||||||
| N0 | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| N+ | 1.66 (1.22-2.25) | <.001 | 3.41 (2.34-4.98) | <.001 | 1.48 (1.02-2.15) | .04 |
| AJCC 8 stage | ||||||
| Stage 1 | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| Stage 2 | 1.34 (0.85-2.10) | .21 | 2.51 (1.45-4.33) | .001 | 1.52 (1.14-2.03) | .005 |
| Stage 3 | 1.57 (1.06-2.31) | .02 | 4.26 (2.50-7.25) | <.001 | 1.62 (1.05-2.48) | .03 |
| Stage 4 | 4.34 (2.73-6.90) | <.001 | 10.4 (4.60-23.6) | <.001 | 2.31 (1.30-4.12) | .005 |
| Treatment | ||||||
| Surgery alone | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| Radiotherapy alone | 1.26 (0.79-2.02) | .33 | 1.51 (0.55-4.12) | .42 | 0.78 (0.37-1.65) | .52 |
| Surgery plus radiotherapy | 0.70 (0.50-0.96) | .03 | 0.99 (0.52-1.90) | .98 | 0.82 (0.56-1.20) | .30 |
| Observation | 1.78 (1.21-2.60) | .003 | 1.01 (0.37-2.70) | .99 | 1.42 (0.82-2.44) | .21 |
Abbreviations: AJCC, American Joint Committee on Cancer; HR, hazard ratio; NE, not estimable.
Table 4. Multivariate Analysis of Survival Outcomes Adjusted by Age, Sex, Immunosuppression, Stage, and Treatment.
| Outcome | Overall survival | Disease-specific survival | Recurrence-free survival | |||
|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |
| Age | 1.04 (1.02-1.06) | <.001 | 1.00 (0.99-1.01) | .98 | 0.99 (0.97-1.02) | .62 |
| Sex | ||||||
| Male | 1.16 (0.80-1.68) | .42 | 1.95 (1.02-3.74) | .04 | 2.06 (1.58-2.67) | <.001 |
| Female | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| Immunosuppression | 2.29 (1.56-3.37) | <.001 | 2.85 (1.98-4.11) | <.001 | 2.82 (1.85-4.29) | <.001 |
| AJCC 8 stage | ||||||
| Stage 1 | [Reference] | NE | [Reference] | NA | [Reference] | NE |
| Stage 2 | 1.35 (0.90-2.01) | .15 | 2.93 (1.58-5.42) | <.001 | 1.66 (1.37-2.01) | <.001 |
| Stage 3 | 2.06 (1.41-3.02) | <.001 | 4.96 (2.16-11.4) | <.001 | 1.76 (1.13-2.74) | .01 |
| Stage 4 | 5.92 (3.98-8.81) | <.001 | 12.3 (3.63-41.8) | <.001 | 2.73 (1.42-5.23) | .003 |
| Treatment | ||||||
| Surgery alone | [Reference] | NE | [Reference] | NE | [Reference] | NE |
| Radiotherapy alone | 1.15 (0.70-1.89) | .58 | 1.24 (0.46-3.32) | .67 | 0.67 (0.35-1.26) | .21 |
| Surgery plus radiotherapy | 0.76 (0.46-1.25) | .28 | 0.66 (0.26-1.71) | .40 | 0.72 (0.44-1.18) | .19 |
| Observation | 1.93 (1.26-2.96) | .003 | 1.80 (0.62-5.25) | .28 | 1.62 (0.96-2.71) | .07 |
Abbreviations: AJCC, American Joint Committee on Cancer; HR, hazard ratio; NE, not estimable.
Surgery with radiotherapy was significantly associated with improved OS compared with surgery alone in univariable analysis. However, with adjustment for age, sex, immunosuppression, and stage, this association was no longer statistically significant (HR, 0.70 [95% CI, 0.50-0.96]; adjusted HR, 0.76 [95% CI, 0.46-1.25]) (Table 3 and Table 4). Patients who received no treatment had worse OS in both univariable and multivariable analyses (HR, 1.78 [95% CI, 1.21-2.60]; adjusted HR, 1.93 [95% CI, 1.26-2.96]).
Kaplan-Meier curves demonstrating survival outcomes by AJCC 8 staging and treatment type are presented in the Figure and eFigure 3 in Supplement 1, respectively. After adjustment for age, sex, and immunosuppression status, OS and DSS were highest among patients treated with surgery and radiotherapy, followed by surgery alone and radiotherapy alone across all stages (eFigure 3 in Supplement 1). Patients treated with radiotherapy (alone or with surgery) had the highest RFS across all stages (eFigure 3 in Supplement 1).
Discussion
Merkel cell carcinoma is an aggressive cutaneous malignant neoplasm. Treatment of MCC is challenging because the disease is rare and studies guiding treatment are limited to small single-institution retrospective studies. In this investigation, we report, to our knowledge, the largest Canada-wide assessment of HNMCC survival outcomes among patients who received various treatments including surgery, radiotherapy, or multimodal therapy.
Survival
Due to differences in patient populations and small sample sizes, HNMCC survival outcomes are variable in the literature. Overall, 5-year OS is reported between 34.4% and 83%, and DSS is reported between 21% and 90%.8,11,12,13,14,15,16 In the absence of large, multicenter randomized clinical trials, the National Cancer Database (NCDB) and the Surveillance, Epidemiology, and End Results (SEER) program have been used to derive large cohorts to further investigate prognostic and survival factors in HNMCC. Based on these databases, 5-year OS estimates for HNMCC are between 41% and 43%, and 5-year DSS is approximately 67.9%.13,16 Survival outcomes were similar in our large Canadian cohort, with a 5-year OS of 42.0% and 5-year DSS of 75.9%. Few studies have reported RFS, but based on our study, approximately 50% of patients will experience recurrence within 5 years.
To our knowledge, the largest study reporting survival outcomes, by Harms et al,6 included 9387 patients with MCC and formed the validation cohort of the AJCC 8 staging system. Within that cohort, 42.6% of patients had HNMCC. Based on these staging criteria, we found that the Canadian experience with HNMCC is comparable to the OS outcomes in the cohort described by Harms et al,6 with slightly worse outcomes for stage I or local disease (eTable 2 and eFigure 2 in Supplement 1).
Patient or Tumor Prognostic Factors
Merkel cell carcinoma is known to predominantly affect older individuals, with median age of diagnosis at 75 to 80 years of age.2 Although we found that increasing age is associated with worse OS, there was no significant association of age with RFS or DSS. Conversely, while sex was not associated with worse OS, men did have worse DSS and RFS compared with women. An analysis of both the NCDB and SEER databases showed increased mortality among men compared with women with MCC, which was associated with greater cancer-specific mortality among men and was even more pronounced among individuals with HNMCC.17 Although the underlying mechanism for this association is unclear, it is postulated that differences in immune system activation between men and women may be associated with differential anticancer immune response.18
Regardless of sex, immunosuppression was also independently associated with all survival outcomes in our study, which has been corroborated by several investigations.19,20,21 Although we were not able to stratify patients according to the nature of immunosuppression, Yusuf et al22 demonstrated that solid organ transplants appear to have worse survival outcomes compared with other means of immunosuppression, including HIV-AIDS and hematologic malignant neoplasms. With increasing rates of organ transplants, and longer life spans with improved treatments for other immunosuppressive conditions, immune status should be considered in the prognostication of patients with HNMCC.
Treatment Patterns
National Comprehensive Cancer Network guidelines for HNMCC recommend treatment for localized tumors to include surgical excision followed by adjuvant radiotherapy or observation, favoring the use of radiotherapy for patients with HNMCC for its potentially limited ability to achieve 1- to 2-cm margins and the risk of false-negative sentinel lymph node biopsy results.23 Furthermore, MCC is a radiosensitive malignant neoplasm, and postoperative radiotherapy has shown improved outcomes, including increased OS and disease-free survival, compared with surgery alone.15,16,24,25,26 In our Canadian cohort, multimodal treatment with surgery and radiotherapy was administered to over one-third of patients with stage I and II diseases and more than 50% of patients with stage III disease. Adjuvant radiotherapy was associated with better OS compared with surgical excision alone on univariable analysis. Although this survival benefit was no longer significant on adjusted multivariable analysis, survival curves demonstrated that patients across all stages who were treated with surgery and adjuvant radiotherapy had improved OS and DSS over time (eFigure 3 in Supplement 1). The lack of statistical significance on adjusted multivariable analysis may be explained by the inclusion of patients with stage IV disease who have poor outcomes regardless of treatment modality or because the study was underpowered to detect a statistically significant difference. The data showed that OS was significantly worse with observation alone compared with surgery, as well as worse survival compared with all other treatments, suggesting that treatment in the form of surgery (with or without radiotherapy) is potentially associated with a survival benefit for patients.
Radiotherapy alone was most frequently used for patients with stage IV disease and was also used for a large proportion of patients with stage I to stage III disease. After adjustment for stage, there was no significant survival difference between radiotherapy alone and surgical intervention alone, suggesting that it may be reasonable to consider the use of radiotherapy alone for patients who refuse surgery or are not suitable surgical candidates.27
There were no statistically significant differences in RFS or DSS between treatment options on multivariate analysis. Further research is required to identify treatment modalities that reduce recurrence of disease to improve outcomes in MCC. Given the association between MCC and immune status, various immune checkpoint inhibitors including the programmed cell death ligand 1 inhibitor drug avelumab and the anti–programmed cell death 1 drugs nivolumab and pembrolizumab have been investigated in metastatic MCC. Due to the time period of this study, patients in this cohort did not receive immunotherapy; however, these immunotherapies are now recommended for metastatic MCC in the National Comprehensive Cancer Network guidelines,23 and avelumab and pembrolizumab are currently being investigated for surgically treated patients with stage I to stage III disease (NCT03271372, NCT03712605, and NCT04291885).
Strengths and Limitations
This study has some strengths. With nearly 400 patients from 10 university centers and 3 provincial cancer registries across Canada, our findings represent the first and largest MCC cohort in Canada as well as one of the largest reported in the literature. Furthermore, the OS, DSS, and RFS curves presented here are specific to patients with HNMCC and stratified according to stage and type of treatment, which can be used to provide pertinent information for a variety of clinical scenarios (eTables 1 and 2 in Supplement 1).
This study also has some limitations. Although there is the potential for referral bias as patients were selected from academic institutions, these findings are likely still generalizable to other populations because an increasing proportion of patients with MCC are evaluated and treated at tertiary care centers.28 Although the Pan-Canadian Merkel Cell Collaborative database collects a large breadth of information, there are limitations and incomplete data for certain variables, including biopsy and sentinel lymph node biopsy pathology, within the patient population with HNMCC, which limited our ability to evaluate the association of these clinical factors with treatment and survival. In addition, we were unable to subdivide the patients who received radiotherapy according to treatment at the nodal site vs the primary site. Finally, there may be unmeasured confounders and selection bias among the patient population that underwent surgery.
Conclusion
To our knowledge, this is the largest assessment of survival outcomes among Canadian patients with HNMCC. Although this cohort study provides insights into the importance of multidisciplinary care as well as the development of novel therapies, the prognosis for patients with MCC remains poor, and, therefore, further research is required to determine optimal treatment by stage of disease.
eTable 1. Demographics
eTable 2. Comparison of Overall Survival Outcomes by Disease Extent and AJCC 8 Staging Between Current Cohort and the AJCC 8 Validation Cohort
eFigure 1. Patient Flow Chart
eFigure 2. 5-Year Survival Outcomes by Disease Extent
eFigure 3. 5-Year Adjusted Overall Survival (OS), Disease Specific Survival (DSS), and Recurrence Free Survival (RFS) by Treatment Type, Stratified by AJCC Stage
Data Sharing Statement
References
- 1.Walsh NM, Cerroni L. Merkel cell carcinoma: a review. J Cutan Pathol. 2021;48(3):411-421. doi: 10.1111/cup.13910 [DOI] [PubMed] [Google Scholar]
- 2.Becker JC, Stang A, DeCaprio JA, et al. Merkel cell carcinoma. Nat Rev Dis Primers. 2017;3:17077. doi: 10.1038/nrdp.2017.77 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Yusuf MB, McKenzie G, Rattani A, et al. Merkel cell carcinoma of the head and neck: epidemiology, pathogenesis, current state of treatment and future directions. Cancers (Basel). 2021;13(14):3506. doi: 10.3390/cancers13143506 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Paulson KG, Park SY, Vandeven NA, et al. Merkel cell carcinoma: current US incidence and projected increases based on changing demographics. J Am Acad Dermatol. 2018;78(3):457-463. doi: 10.1016/j.jaad.2017.10.028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Schadendorf D, Lebbé C, Zur Hausen A, et al. Merkel cell carcinoma: epidemiology, prognosis, therapy and unmet medical needs. Eur J Cancer. 2017;71:53-69. doi: 10.1016/j.ejca.2016.10.022 [DOI] [PubMed] [Google Scholar]
- 6.Harms KL, Healy MA, Nghiem P, et al. Analysis of prognostic factors from 9387 Merkel cell carcinoma cases forms the basis for the new 8th edition AJCC staging system. Ann Surg Oncol. 2016;23(11):3564-3571. doi: 10.1245/s10434-016-5266-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Morand GB, Madana J, Da Silva SD, Hier MP, Mlynarek AM, Black MJ. Merkel cell carcinoma of the head and neck: poorer prognosis than non–head and neck sites. J Laryngol Otol. 2016;130(4):393-397. doi: 10.1017/S0022215116000153 [DOI] [PubMed] [Google Scholar]
- 8.Haerle SK, Shiau C, Goldstein DP, et al. Merkel cell carcinoma of the head and neck: potential histopathologic predictors. Laryngoscope. 2013;123(12):3043-3048. doi: 10.1002/lary.24233 [DOI] [PubMed] [Google Scholar]
- 9.Fritsch VA, Camp ER, Lentsch EJ. Sentinel lymph node status in Merkel cell carcinoma of the head and neck: not a predictor of survival. Head Neck. 2014;36(4):571-579. doi: 10.1002/hed.23334 [DOI] [PubMed] [Google Scholar]
- 10.Ahmad TR, Vasudevan HN, Lazar AA, et al. Should sentinel lymph node biopsy status guide adjuvant radiation therapy in patients with Merkel cell carcinoma? Adv Radiat Oncol. 2021;6(6):100764. doi: 10.1016/j.adro.2021.100764 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lok B, Khan S, Mutter R, et al. Selective radiotherapy for the treatment of head and neck Merkel cell carcinoma. Cancer. 2012;118(16):3937-3944. doi: 10.1002/cncr.26738 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Bowe CM, Bean T, Loke R, et al. Merkel cell carcinoma of the head and neck in the south-east of England. Br J Oral Maxillofac Surg. 2021;59(10):1280-1286. doi: 10.1016/j.bjoms.2021.06.009 [DOI] [PubMed] [Google Scholar]
- 13.Go CC, Kim DH, Briceño CA. A SEER analysis of survival and prognostic factors in merkel cell carcinoma of the head and neck region. Int J Oral Maxillofac Surg. 2022;51(3):314-322. doi: 10.1016/j.ijom.2021.05.014 [DOI] [PubMed] [Google Scholar]
- 14.Morand G, Vital D, Pézier T, et al. Merkel cell carcinoma of the head and neck: a single institutional experience. J Skin Cancer. 2013;2013:325086. doi: 10.1155/2013/325086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Clark JR, Veness MJ, Gilbert R, O’Brien CJ, Gullane PJ. Merkel cell carcinoma of the head and neck: is adjuvant radiotherapy necessary? Head Neck. 2007;29(3):249-257. doi: 10.1002/hed.20510 [DOI] [PubMed] [Google Scholar]
- 16.Chen MM, Roman SA, Sosa JA, Judson BL. The role of adjuvant therapy in the management of head and neck Merkel cell carcinoma: an analysis of 4815 patients. JAMA Otolaryngol Head Neck Surg. 2015;141(2):137-141. doi: 10.1001/jamaoto.2014.3052 [DOI] [PubMed] [Google Scholar]
- 17.Tam M, Luu M, Barker CA, et al. Improved survival in women versus men with Merkel cell carcinoma. J Am Acad Dermatol. 2021;84(2):321-329. doi: 10.1016/j.jaad.2020.02.034 [DOI] [PubMed] [Google Scholar]
- 18.Klein SL, Flanagan KL. Sex differences in immune responses. Nat Rev Immunol. 2016;16(10):626-638. doi: 10.1038/nri.2016.90 [DOI] [PubMed] [Google Scholar]
- 19.Yusuf M, Gaskins J, May ME, et al. Immune status and the efficacy of adjuvant radiotherapy for patients with localized Merkel cell carcinoma of the head and neck. Clin Transl Oncol. 2020;22(11):2009-2016. doi: 10.1007/s12094-020-02338-2 [DOI] [PubMed] [Google Scholar]
- 20.Tarantola TI, Vallow LA, Halyard MY, et al. Prognostic factors in Merkel cell carcinoma: analysis of 240 cases. J Am Acad Dermatol. 2013;68(3):425-432. doi: 10.1016/j.jaad.2012.09.036 [DOI] [PubMed] [Google Scholar]
- 21.Tseng YD, Nguyen MH, Baker K, et al. Effect of patient immune status on the efficacy of radiation therapy and recurrence-free survival among 805 patients with Merkel cell carcinoma. Int J Radiat Oncol Biol Phys. 2018;102(2):330-339. doi: 10.1016/j.ijrobp.2018.05.075 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Yusuf MB, Gaskins J, Rattani A, et al. Immune status in Merkel cell carcinoma: relationships with clinical factors and independent prognostic value. Ann Surg Oncol. 2021;28(11):6154-6165. doi: 10.1245/s10434-021-09944-6 [DOI] [PubMed] [Google Scholar]
- 23.National Comprehensive Cancer Network . Merkel cell carcinoma. NCCN Guidelines Version 2. National Comprehensive Cancer Network ; 2022. [Google Scholar]
- 24.Balakrishnan V, Berry S, Stew B, Sizeland A. Benefits of combined modality treatment of Merkel cell carcinoma of the head and neck: single institution experience. J Laryngol Otol. 2013;127(9):908-916. doi: 10.1017/S0022215113001862 [DOI] [PubMed] [Google Scholar]
- 25.Howle JR, Hughes TM, Gebski V, Veness MJ. Merkel cell carcinoma: an Australian perspective and the importance of addressing the regional lymph nodes in clinically node-negative patients. J Am Acad Dermatol. 2012;67(1):33-40. doi: 10.1016/j.jaad.2011.07.029 [DOI] [PubMed] [Google Scholar]
- 26.Veness MJ, Morgan GJ, Gebski V. Adjuvant locoregional radiotherapy as best practice in patients with Merkel cell carcinoma of the head and neck. Head Neck. 2005;27(3):208-216. doi: 10.1002/hed.20134 [DOI] [PubMed] [Google Scholar]
- 27.Mortier L, Mirabel X, Fournier C, Piette F, Lartigau E. Radiotherapy alone for primary Merkel cell carcinoma. Arch Dermatol. 2003;139(12):1587-1590. doi: 10.1001/archderm.139.12.1587 [DOI] [PubMed] [Google Scholar]
- 28.Jacobs D, Olino K, Park HS, et al. Primary treatment selection for clinically node-negative Merkel cell carcinoma of the head and neck. Otolaryngol Head Neck Surg. 2021;164(6):1214-1221. doi: 10.1177/0194599820967001 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Demographics
eTable 2. Comparison of Overall Survival Outcomes by Disease Extent and AJCC 8 Staging Between Current Cohort and the AJCC 8 Validation Cohort
eFigure 1. Patient Flow Chart
eFigure 2. 5-Year Survival Outcomes by Disease Extent
eFigure 3. 5-Year Adjusted Overall Survival (OS), Disease Specific Survival (DSS), and Recurrence Free Survival (RFS) by Treatment Type, Stratified by AJCC Stage
Data Sharing Statement