Abstract
Introduction
The National Comprehensive Cancer Network (NCCN) recommends excision of the primary tumor using 1–2-cm surgical margins and sentinel lymph node biopsy (SLNB) as the initial management of early-stage Merkel cell carcinoma (MCC). However, there is no clear consensus on the appropriate size of the surgical margins and/or the use of Mohs micrographic surgery (MMS). Our aim was to demonstrate that, independent of the type of surgery, obtaining negative surgical margins is associated with enhanced overall survival (OS).
Methods
A retrospective study was performed using early-stage MCC patients from the National Cancer Database (NCDB) who were diagnosed between 2004 and 2020 and underwent surgical excision (SE) of their primary tumor. Patients were stratified into three groups based on the surgical treatment they received: < 1 cm excision margin, ≥ 1 cm excision margin, or MMS. OS was assessed with Kaplan–Meier curves, log-rank tests, and multivariable risk-adjusted Cox proportional-hazards regression.
Results
Of the 4,607 patients included in this study; 53% underwent SE of ≥ 1 cm (n = 2,474), 41% had SE < 1 cm (n = 1,905), and the remainder had MMS (n = 228). 75% of patients had an SLNB, and 56% received adjuvant radiation to the primary site and/or nodal basin. While no difference in OS was observed between the three surgical treatments, negative surgical margins (hazard ratio (HR) 0.78; 95% confidence interval (CI) 0.65–0.94) and receipt of radiation to the primary site (HR 0.82; 95% CI 0.73–0.92) were both independently associated with improved OS.
Conclusion
Achieving negative surgical margins is associated with improved OS in MCC, suggesting that MMS and conventional excision are both suitable approaches for the treatment of primary MCC.
Supplementary Information
The online version contains supplementary material available at 10.1007/s13555-025-01345-x.
Keywords: Merkel cell carcinoma, Surgical margins, Negative surgical margins, NCDB, Survival
Key Summary Points
| The National Comprehensive Cancer Network (NCCN) recommends excision of the primary tumor using 1–2-cm surgical margins and sentinel lymph node biopsy (SLNB) as the initial management of early-stage Merkel cell carcinoma (MCC) patients. There is no clear consensus on the appropriate size of the surgical margins and/or the use of Mohs micrographic surgery (MMS). |
| We hypothesized that, independent of the type of surgery, obtaining negative surgical margins is associated with overall survival (OS). |
| There is no difference in OS between conventional excision (< 1 cm or ≥ 1 cm excision margin) and MMS. Negative surgical margins (HR 0.78; 95% CI 0.65–0.94) and receipt of radiation to the primary site (hazard ratio (HR) 0.82; 95% confidence interval (CI) 0.73–0.92) were both independently associated with improved OS. |
| MMS and conventional excision are both suitable approaches for the treatment of primary MCC. |
Introduction
Merkel cell carcinoma (MCC) is an uncommon, aggressive neuroendocrine tumor that is rapidly increasing in incidence. There are expected to be more than 3000 new cases in the US in 2025 [1–3].
MCC manifests as a fast-growing, reddish to pinkish, cutaneous or subcutaneous nodule that predominantly appears in areas exposed to sunlight. MCC primarily affects elderly, non-Hispanic white individuals, with 40–50% of primary tumors arising in the head and neck region [4–7]. Although most patients are diagnosed at early stages with clinically node-negative disease, the 5-year overall survival (OS) of patients with MCC is 62.8% and 34.8–54.6% for American Joint Committee on Cancer (AJCC) stage I and II disease, respectively. 26–30% of patients have occult nodal metastases at diagnosis, and 8% have distant metastases at initial presentation [6, 8, 9].
The standard of care for MCC patients involves surgical excision (SE) of the primary tumor, usually accompanied by an sentinel lymph node biopsy (SLNB). The National Comprehensive Cancer Network (NCCN) recommends that surgeons performing an SE should employ 1–2-cm measured surgical margins [3, 10, 11]. However, patients may undergo Mohs micrographic surgery (MMS) to ensure that the smallest quantity of viable tissue is removed from sensitive regions such as the head and neck. Patients with adverse risk factors or tumor-positive sentinel lymph nodes (+ SLN) often receive postoperative adjuvant radiotherapy to the site of their primary tumor and/or affected nodal basins [2, 6, 8, 12–14].
A number of studies have attempted to determine the optimal size of surgical margins and the clinical utility of adjuvant radiotherapy [5, 10, 11, 15]. Mixed results have left clinicians without a clear consensus on the best treatment for MCC patients. Additionally, those studies typically excluded patients who undergo MMS.
Several studies have reported similar or better OS for patients who received MMS—which relies on obtaining clear surgical margins—compared to SE [16–18]. Thus, we hypothesized that the pathologic status of the excision margins is more important for OS in MCC patients than the measured size of the margins. To assess the optimal surgical approach for early-stage MCC patients and evaluate the use of adjuvant radiotherapy, we conducted a retrospective study using the National Cancer Database (NCDB). This study aimed to compare OS among patients based on their surgical treatment: MMS versus SE with different surgeon-measured margin sizes and the receipt of adjuvant radiation treatment versus surgery alone.
Methods
Study Design and Patients
We queried the NCDB between 2004 to 2020 and identified 15,416 MCC patients between 18 and 80 years old with histologically confirmed MCC (International Classification of Diseases for Oncology histology code 8247). We included patients with lesions located on the head/neck region, trunk, or extremities (PRIMARY_SITE codes 44.0, 44.1, 44.2, 44.3, 44.4, 44.5, 44.6, 44.7) with an AJCC T stage of 1, 2, or 3 (cT1, cT2, cT3) without clinical nodal involvement (cN0) or distant metastases (cM0). Patients were excluded from the analyses if they had no surgery, if they had an amputation, or if the size of their primary tumor was unknown. They were also excluded if OS data were missing, if they did not receive surgery within 180 days of diagnosis, or if they received systemic adjuvant therapy, immunotherapy, chemotherapy, or radiation after 180 days post-surgery. Our final cohort consisted of 4,607 patients (Fig. 1).
Fig. 1.
Patient selection flow chart; NCDB - National Cancer Database
Patients were divided into three groups based on the surgical management of the primary tumor: (1) SE with margins < 1 cm (surgical codes 20, 27, 30, 31, 32, 33), (2) SE with margins ≥ 1 cm (surgical codes 45, 46, 47), or (3) MMS (surgical codes 34, 35, 36). Clinical excision margin size was determined from the surgical procedure for the primary site by the surgeon (defined by the code RX_SUMM_SURG_PRIM_SITE in the NCDB).
Negative surgical margins (R0 resection) are defined in NCDB as microscopically negative for residual tumor (defined by RX_SUMM_SURGICAL_MARGINS, code 0). Patients with residual tumor in their surgical margins (= positive surgical margins) were identified with the codes 1, 2, and 3.
The management of the nodal basin was based on the type of surgery the patient had (defined by the code RX_SUMM_SCOPE_REG_LN_SUR): SLNB (code 1), no regional lymph node surgery (code 0), or unknown (code 9). We defined three groups based on receipt of post-operative radiation (RAD_LOCATION_OF_RX): did not receive radiation (code 0), received radiation (codes 1, 2, 3, 4, and 8), and unknown radiation status (code 9). The site of radiation was defined using the criteria PHASE_I_RT_VOLUME, PHASE_II_RT_VOLUME, and PHASE_III_RT_VOLUME as primary tumor (codes 10–97), LN basin (codes 1–9), unknown location (code 98), or the patient did not receive radiation (code 99). PHASE_I_RT_TO_LN, PHASE_II_RT_TO_LN, and PHASE_III_RT_TO_LN were used to identify patients who received radiation to their LN basin (codes 1–8) in addition to their primary tumor.
Patients’ baseline health status was assessed by analyzing the average age of the patients in each group and through the Charlson–Deyo Comorbidity Scale, which incorporates 17 different comorbidities (excluding cancer) and reflects the severity of the specific comorbidities of the patients.
This study was deemed exempt by the guidelines set forth by the institutional review board. This study did not require ethical approval from an institutional review board (IRB) as it solely involved publicly available data with no identifiable personal information.
Statistical Analysis
The patients’ clinicopathological features and treatment characteristics were compared using chi-squared tests and ANOVA analyses for categorical and continuous variables, respectively. Multivariable logistic regression analyses were performed to examine factors associated with receiving post-operative radiation, achieving negative surgical margins, and undergoing SLNB.
OS was calculated from date of diagnosis to the time of death or the date censored at last known follow-up and compared between patient groups using the Kaplan–Meier (KM) method with the log-rank test [19]. We present 6-month landmark results that account for differences in time from diagnosis to surgery and the associated immortal time bias [20]. To examine the association between the surgical treatment groups and OS, preliminary univariate Cox proportional-hazards regressions on all factors were performed, followed by multivariable risk-adjusted Cox proportional-hazards regression.
All statistical tests were two-sided, and significance was set at p < 0.05. The multivariable regressions retained the main grouping variables and any other factors that had significant (p < 0.05) associations with the outcome in univariable analysis. All statistical analyses were performed using R software, version 4.2.2.
Results
Patient Characteristics
4607 patients between 18 and 80 years old who were diagnosed between 2004 and 2020 with early-stage (T1–2–3/cN0/cM0) MCC were included in our dataset. Patients were predominantly Caucasian (94.7%; n = 4331), male (61.4%; n = 2828), and ≥ 60 years old (86.7%; n = 3992) with no comorbidities (Charlson score = 0; 72.1%; n = 3323). Most primaries were located on extremities (49.8%) or in the head and neck region (40.0%); they were occasionally located on the trunk (10.2%) (Table 1).
Table 1.
Patients’ clinicopathological characteristics
| All (n = 4607) | Mohs surgery (n = 228) | Surgical excision < 1 cm (n = 1905) | Surgical excision ≥ 1 cm (n = 2474) | p value | |
|---|---|---|---|---|---|
| Median age (SD) | 69.36 (8.44) | 69.87 (8.64) | 69.67 (8.26) | 69.08 (8.55) | 0.047 |
| Age group | 0.147 | ||||
| 18–59 | 615 (13.35%) | 25 (10.96%) | 238 (12.49%) | 352 (14.23%) | |
| 60–80 | 3992 (86.65%) | 203 (89.04%) | 1667 (87.51%) | 2122 (85.77%) | |
| Gender | 0.538 | ||||
| Male | 2828 (61.38%) | 132 (57.89%) | 1172 (61.52%) | 1524 (61.6%) | |
| Female | 1779 (38.62%) | 96 (42.11%) | 733 (38.48%) | 950 (38.4%) | |
| Charlson score | < 0.001 | ||||
| 0 | 3323 (72.13%) | 193 (84.65%) | 1356 (71.18%) | 1774 (71.71%) | |
| 1 | 852 (18.49%) | 22 (9.65%) | 346 (18.16%) | 484 (19.56%) | |
| 2 + | 432 (9.38%) | 13 (5.7%) | 203 (10.66%) | 216 (8.73%) | |
| Clinical T stage | 0.007 | ||||
| cT1 | 3353 (72.78%) | 188 (82.46%) | 1393 (73.12%) | 1772 (71.62%) | |
| cT2 | 1098 (23.83%) | 38 (16.67%) | 446 (23.41%) | 614 (24.82%) | |
| cT3 | 156 (3.39%) | 2 (0.88%) | 66 (3.46%) | 88 (3.56%) | |
| Primary site | < 0.001 | ||||
| Trunk | 469 (10.18%) | 6 (2.63%) | 174 (9.13%) | 289 (11.68%) | |
| Extremities | 2294 (49.79%) | 72 (31.58%) | 875 (45.93%) | 1347 (54.45%) | |
| Head and neck | 1844 (40.03%) | 150 (65.79%) | 856 (44.93%) | 838 (33.87%) | |
| Pathological T stage | 0.042 | ||||
| T0 | 45 (1.07%) | 4 (2.06%) | 14 (0.81%) | 27 (1.19%) | |
| T1 | 2996 (71.38%) | 155 (79.9%) | 1251 (72.15%) | 1590 (70.07%) | |
| T2 | 935 (22.28%) | 32 (16.49%) | 379 (21.86%) | 524 (23.09%) | |
| T3 | 142 (3.38%) | 2 (1.03%) | 55 (3.17%) | 85 (3.75%) | |
| T4 | 79 (1.88%) | 1 (0.52%) | 35 (2.02%) | 43 (1.9%) | |
| Tx/unknown | 410 | 34 | 171 | 205 | |
| Pathological N stage | 0.092 | ||||
| N0 | 2795 (78.53%) | 121 (83.45%) | 1098 (78.09%) | 1576 (78.49%) | |
| N1 | 725 (20.37%) | 22 (15.17%) | 286 (20.34%) | 417 (20.77%) | |
| N2/3 | 39 (1.1%) | 2 (1.38%) | 22 (1.56%) | 15 (0.75%) | |
| Nx/unknown | 1048 | 83 | 499 | 466 | |
| Lymphovascular invasion | 0.167 | ||||
| No | 1905 (69.32%) | 74 (77.08%) | 828 (68.26%) | 1003 (69.7%) | |
| Yes | 843 (30.68%) | 22 (22.92%) | 385 (31.74%) | 436 (30.3%) | |
| Unknown | 1859 | 132 | 692 | 1035 | |
2474 (53.7%) patients had an SE with surgical margins ≥ 1 cm, 1905 (41.3%) had an SE with surgical margins < 1 cm, and 228 (5.0%) patients underwent MMS. MMS was more often used for primaries of the head and neck region (65.8%). The majority of patients (92.9%) achieved negative surgical margins. 75% of patients underwent an SLNB at the time of primary tumor excision: 71.1%, 79.8%, and 56.0% of patients who had an SE < 1 cm, an SE ≥ 1 cm, and MMS, respectively (p < 0.001). 22.3% of patients who underwent SLNB had a + SLN. 56.2% of patients received post-operative radiation therapy. Radiation was most commonly administered to the primary tumor site (40.6%); only 90 patients (2.1%) received radiation to their lymph node basin alone. A greater proportion of patients in the MMS group (67.5%) received radiation therapy compared to patients in the SE groups (55.7% and 55.5% for < 1 cm and ≥ 1 cm, respectively) (p < 0.001) (Table 2).
Table 2.
Surgical and therapeutic management of the patients
| All (n = 4607) | Mohs surgery (n = 228) | Surgical excision < 1 cm (n = 1905) | Surgical excision ≥ 1 cm (n = 2474) | p value | |
|---|---|---|---|---|---|
| Surgical margin status | < 0.001 | ||||
| Negative | 4279 (92.88%) | 219 (96.05%) | 1685 (88.45%) | 2375 (96%) | |
| Positive | 328 (7.12%) | 9 (3.95%) | 220 (11.55%) | 99 (4%) | |
| Underwent SLNB | < 0.001 | ||||
| No | 1149 (24.97%) | 100 (44.05%) | 549 (28.86%) | 500 (20.22%) | |
| Yes | 3453 (75.03%) | 127 (55.95%) | 1353 (71.14%) | 1973 (79.78%) | |
| Unknown | 5 | 1 | 3 | 1 | |
| SLN status | 0.444 | ||||
| Negative | 2639 (77.75%) | 102 (82.26%) | 1026 (77.26%) | 1511 (77.81%) | |
| Positive | 755 (22.25%) | 22 (17.74%) | 302 (22.74%) | 431 (22.19%) | |
| Unknown status | 59 | 3 | 25 | 31 | |
| Unknown SLNB | 1154 | 101 | 552 | 501 | |
| Receipt of radiation | 0.001 | ||||
| No | 1962 (43.82%) | 824 (44.35%) | 72 (32.43%) | 1066 (44.47%) | |
| Yes | 2515 (56.18%) | 1034 (55.65%) | 150 (67.57%) | 1331 (55.53%) | |
| Unknown | 130 | 47 | 6 | 77 | |
| Location of radiation | 0.001 | ||||
| None | 1991 (45.61%) | 73 (33.64%) | 1081 (46.36%) | 837 (46.09%) | |
| Primary tumor | 1772 (40.6%) | 100 (46.08%) | 951 (40.78%) | 721 (39.7%) | |
| Lymph node basin | 90 (2.06%) | 7 (3.23%) | 50 (2.14%) | 33 (1.82%) | |
| Both primary and basin | 512 (11.73%) | 37 (17.05%) | 250 (10.72%) | 225 (12.39%) | |
| NA/unknown | 242 | 11 | 142 | 89 | |
SLNB sentinel lymph node biopsy; SLN sentinel lymph node
The SE groups (< 1 cm and ≥ 1 cm) were similar in respect to distribution of T stage (cT1 vs. cT2 vs. cT3), whereas MMS had a greater percentage of patients with cT1 lesions (Table 1). The majority of patients (72.8%) had a clinical T1 lesion. Interestingly, cT1 patients were older (69.7 ± 8.2) than cT2 and cT3 patients (68.7 ± 8.8 and 67.3 ± 9.8, respectively) (p < 0.001). More cT1 patients had negative surgical margins (94.9%) compared to cT2 and cT3 patients (88.3% and 81.4%, respectively; p < 0.001), and more cT1 patients underwent SLNB (77.0%) compared to cT2 and cT3 patients (70.5% and 65.4%, respectively; p < 0.001). Fewer cT1 patients had + SLN (19.3%) compared to cT2 and cT3 patients (31.4% and 27.0%, respectively; p < 0.001), and fewer cT1 patients (53.0%) received radiation compared to cT2 and cT3 patients (65.6% and 56.0%, respectively; p < 0.001) (Supplemental Tables 1 and 2).
Patient clinicopathological features and treatment characteristics for the three groups are summarized in Tables 1 and 2 and Supplemental Tables 1 and 2.
Survival Analysis
The median OS (95% CI) for the entire cohort was over 10 years (121.4 months) [95% CI 108.9–140.2], with a median OS of 131.8 [111.4–151.7], 130.2 [121.0–142.1], and 115.3 [103.4–infinity] months for the < 1 cm SE, ≥ 1 cm SE, and MMS groups, respectively. There was no difference (p = 0.88) in OS based on the type of surgery the patients received (Fig. 2).
Fig. 2.
Overall survival according to surgery of the primary tumor. Comparison of patients who had Mohs micrographic surgery (n = 228; red), surgical excision with < 1 cm margins (n = 1905; blue), and surgical excision with ≥ 1 cm margins (n = 2474; green); SE - surgical excision
On multivariable analysis, older age (≥ 60 relative to 18–59; HR 2.01, 95% CI 1.66–2.44), presence of comorbidities (Charlson ≥ 1; HR 1.48, 95% CI 1.3–1.68), increasing tumor size (HR 1.07, 95% CI 1.01–1.13), + SLN (HR 1.76, 95% CI 1.52–2.04), and lymphovascular invasion (LVI) (HR 1.17, 95% CI 1.0–1.37) were significantly associated with worse OS. Factors significantly associated with improved OS on multivariable analysis included female gender (HR 0.64, 95% CI 0.57–0.72), primary tumors on the extremities (compared to trunk; HR 0.76, 95% CI 0.64–0.9), negative surgical margins (HR 0.78, 95% CI 0.65–0.94), and receipt of radiation to the primary tumor site (HR 0.82, 95% CI 0.73–0.92) (Table 3).
Table 3.
Variables associated with overall survival in patients with Merkel cell carcinoma
| Variable | Univariable | Multivariable | ||||||
|---|---|---|---|---|---|---|---|---|
| HR | conf.low | conf.high | p value | HR | conf.low | conf.high | p value | |
| Surgical procedure | ||||||||
| SE < 1 cm | ref | ref | ||||||
| Mohs Surgery | 0.95 | 0.75 | 1.19 | 0.64 | 1.01 | 0.8 | 1.28 | 0.906 |
| SE ≥ 1 cm | 0.95 | 0.86 | 1.05 | 0.34 | 1.07 | 0.96 | 1.19 | 0.204 |
| Age | ||||||||
| 18–59 | ref | ref | ||||||
| 60–80 | 2.41 | 2.01 | 2.9 | < 0.001 | 2.04 | 1.69 | 2.47 | < 0.001 |
| Gender | ||||||||
| Male | ref | ref | ||||||
| Female | 0.61 | 0.55 | 0.68 | < 0.001 | 0.64 | 0.57 | 0.71 | < 0.001 |
| Charlson comorbidity | ||||||||
| 0 | ref | ref | ||||||
| 1 | 1.43 | 1.26 | 1.61 | < 0.001 | 1.47 | 1.29 | 1.66 | < 0.001 |
| 2 + | 2.22 | 1.91 | 2.58 | < 0.001 | 2.27 | 1.95 | 2.64 | < 0.001 |
| Primary site | ||||||||
| Trunk | ref | ref | ||||||
| Upper/lower extremity | 0.64 | 0.55 | 0.75 | < 0.001 | 0.76 | 0.65 | 0.9 | < 0.001 |
| Head/neck | 0.9 | 0.77 | 1.06 | 0.201 | 1.02 | 0.86 | 1.21 | 0.806 |
| Surgical margins | ||||||||
| Negative (R0) | ref | ref | ||||||
| Positive | 1.78 | 1.51 | 2.09 | < 0.001 | 1.38 | 1.16 | 1.64 | < 0.001 |
| Result of SLNB | ||||||||
| Negative | ref | ref | ||||||
| No/unknown | 2.18 | 1.96 | 2.44 | < 0.001 | 1.88 | 1.67 | 2.12 | < 0.001 |
| Positive | 1.96 | 1.72 | 2.25 | < 0.001 | 1.72 | 1.49 | 1.99 | < 0.001 |
| Unknown | 1.19 | 0.73 | 1.92 | 0.488 | 1.14 | 0.71 | 1.85 | 0.586 |
| Receipt of radiation | ||||||||
| No | ref | ref | ||||||
| Yes | 0.91 | 0.83 | 1.01 | 0.077 | 0.8 | 0.72 | 0.88 | < 0.001 |
| Unknown | 1.38 | 1.07 | 1.8 | 0.014 | 1.14 | 0.87 | 1.48 | 0.336 |
| Location of radiation | ||||||||
| None | ref | ref | ||||||
| Primary | 0.83 | 0.74 | 0.93 | < 0.001 | 0.77 | 0.68 | 0.86 | < 0.001 |
| Lymph node basin | 1.09 | 0.77 | 1.56 | 0.622 | 0.77 | 0.53 | 1.1 | 0.15 |
| Primary and lymph nodes | 1.25 | 1.07 | 1.46 | 0.006 | 0.88 | 0.74 | 1.04 | 0.137 |
| Unknown | 1.07 | 0.86 | 1.32 | 0.545 | 0.96 | 0.78 | 1.19 | 0.727 |
| Lymphovascular invasion | ||||||||
| No | ref | ref | ||||||
| Yes | 1.37 | 1.19 | 1.58 | < 0.001 | 1.19 | 1.02 | 1.38 | 0.022 |
| Unknown | 1.11 | 0.99 | 1.24 | 0.075 | 1.01 | 0.89 | 1.16 | 0.833 |
SE surgical excision; SLNB sentinel lymph node biopsy
Patients with negative surgical margins had a longer median OS (132.1 months [124.2–141.1]) compared to patients with positive surgical margins (90.3 months [57.4–153.0]) (p < 0.0001; Fig. 3).
Fig. 3.
Overall survival based on margin status. Comparisons of patients who had negative surgical margins (n = 328) and positive surgical margins (n = 4279) of resection after excision of the primary tumor
There was no difference in median OS between patients who received radiation (131.8 months [121.0–142.1]) and those who did not (131.6 months [118.4–161.9]) (p = 0.57) (Supplemental Fig. 1).
Predictors of Negative Resection Margin
Multivariable analysis demonstrated that patients who underwent an SE ≥ 1 cm or MMS were significantly more likely to have negative surgical margins than patients who had a < 1 cm SE (adjusted odds ratio [aOR]: SE ≥ 1 cm: 2.91, 95% CI 2.26–3.77; MMS: 3.79, 95% CI 1.99–8.19). On multivariable analysis, pathological T stage > 2 (aOR 0.52, 95% CI 0.32–0.86), presence of + SLN (aOR 0.44, 95% CI 0.28–0.67), increasing primary tumor size (aOR 0.72, 95% CI 0.64–0.82), and tumor located in the head and neck region (compared to the trunk; aOR 0.44, 95% CI 0.28–0.67) were significantly associated with a higher likelihood of positive surgical margins (all p < 0.005; Supplemental Table 5).
Predictors of Receipt of SLNB
Having an SE with surgical margins ≥ 1 cm (aOR 1.72, 95% CI 1.48–2.0) and a primary tumor located on an extremity (compared to the trunk; aOR 1.84, 95% CI 1.44–2.33) were significantly associated with having an SLNB. On the other hand, MMS (compared to a < 1 cm SE; aOR 0.64, 95% CI 0.48–0.87), older age (≥ 60 compared to 18–59; aOR 0.56, 95% CI 0.45–0.71), female gender (aOR 0.77, 95% CI 0.66–0.88), primary tumor located in the head and neck region (compared to the trunk; aOR 0.64, 95% CI 0.5–0.81), and larger tumor size (aOR 0.85, 95% CI 0.79–0.92) were significantly associated with not having an SLNB or nodal evaluating surgery (Supplemental Table 3).
Predictors of Receipt of Radiation
Undergoing MMS (aOR 2.03, 95% CI 1.5–2.76), having positive surgical margins (aOR 1.65, 95% CI 1.27–2.16), the presence of a + SLN (aOR 2.16, 95% CI 1.79–2.62), increasing primary tumor size (aOR 1.13, 95% CI 1.05–1.2), and the presence of LVI (aOR 1.52, 95% CI 1.27–1.82) were significantly associated with the receipt of radiation (Supplemental Table 4).
Discussion
Despite several studies examining the surgical management of early-stage MCC primary tumors, there is no consensus on the best approach for this disease [10, 11, 15–23]. Moreover, although MMS is commonly used by dermatologists, especially for head and neck MCC, this surgical procedure has not been included in most studies [7, 10, 11, 21, 22, 24]. To improve on these studies, we elected to compare the different surgical treatments of early-stage MCC—based on the size of the excision margins—to the use of MMS.
This study compares the surgical management of patients with early-stage MCC and demonstrates that the type of surgery (SE vs. MMS) is not related to OS (Fig. 2) but that achieving negative surgical margins is (Fig. 3). Using the SEER database, Moore et al. described similar results showing no survival differences between patients who received MMS and those who had wide local excision (WLE) [25]. However, this study only included WLE with surgical margins ≥ 1 cm and patients older than 80 years old, which potentially creates a bias due to non-MCC deaths. Additionnally, these studies combined patients with primary tumors located to the trunk and extremities which are known to have different OS [26]. Furthermore, they separated patients with primary tumors located on the scalp and neck from those located on the face. Andruska et al. reported opposite results, showing that wider surgical margins (≥ 1 cm) for MCC were associated with improved OS. However, their study excluded patients who underwent MMS and included patients older than 80 years old [10]. We elected to include patients who received MMS to gain insight into how this treatment compares to narrow and wide excision margins, and we excluded patients older than 80 years old because we observed steep survival curves for them (data not shown) in the initial analysis of this group and to avoid bias due to non-MCC deaths. Other studies have also illustrated that the status of the surgical margins is more prognostic for survival than margin size. Singh et al. showed that neither SE nor MMS was associated with OS but that patients with a positive resection had much worse 3-year OS (40.3% vs. 66.3%; p < 0.0001) [17]. Similarly, Jaouen et al. demonstrated for a small cohort (n = 214) that margin size was not prognostic of survival but that achieving negative surgical margins was associated with significantly improved patient outcomes [15].
Although most MCCs are excised using conventional excisions, MMS appears to be an effective treatment with several advantages [16, 17]. Considering that MCCs appear frequently in the head and neck region, MMS could serve as an effective approach for tumors located where extensive surgery might not be possible due to tight anatomical areas and esthetic concerns. Perez et al. found that the SE margin sizes did not impact survival but that larger margins significantly increased the likelihood of flap or skin graft [11]. Furthermore, studies comparing the recurrence rates of patients treated with SE and MMS have found that MMS leads to a lower recurrence rate (5–17%) than SE (18–50%), most likely due to MMS’s capability to nearly guarantee negative surgical margins [18]. In our study, MMS was predominantly used in T1 lesions, and many received post-operative radiation, making it harder to compare MMS to conventional excision, which may help explain the superior results of MMS to conventional excision.
In our cohort, 75% of patients had an SLNB and 22.3% had a + SLN. Patients with one or more + SLNs had a significantly worse OS than patients with no nodal metastases. The high number of patients who underwent an SLNB is a strength of this study as it differs from many papers on MCC, where either only a minority had an SLNB or only patients with a tumor-negative sentinel node were evaluated [10, 16]. Additionally, the high rate of SLNB use makes the results of this study more generalizable and applicable to real-world scenarios [7, 10, 16].
In our study, 56.2% of patients received adjuvant radiotherapy, most often on their primary tumor. Receiving radiotherapy to the primary tumor site was significantly associated with improved OS. Patients with positive surgical margins, + SLNs, larger primary tumors, and the presence of LVI were more likely to have radiotherapy treatment. We believe that radiotherapy increased the OS of these patients to levels comparable with those of patients who had no adverse risk factors. Our findings align with other studies suggesting that adjuvant radiotherapy is an effective treatment to improve survival in MCC patients [10, 27].
This study has several limitations. First, being an NCDB retrospective study, only OS can be assessed. Second, the NCDB only captures information from Commission on Cancer accredited hospitals, missing MMS surgeries done in outpatient facilities and resulting in a small subset of patients undergoing MMS. Third, most patients had negative surgical margins, making the subset of patients with positive resections small and less powerful statistically. Although there are potential biases in the NCDB, we tried to address them by removing patients over the age of 80.
Conclusion
In conclusion, this study shows that the main criterion associated with improved OS in MCC patients is achieving negative surgical margins at the primary site, independent of the type of surgery. This suggests that MMS is equivalent to, if not better than, narrow and wide local excision for the surgical management of MCC primary tumors, especially for tumors located in the head and neck region, where wide surgical margins cannot always be used.
Supplementary Information
Below is the link to the electronic supplementary material.
Author Contributions
Acquisition of the data: Arthur W. Cowman, Kristel Lourdault, Douglas Hanes, Sean Nassoiy. Analysis of the data: Arthur W. Cowman, Kristel Lourdault, Douglas Hanes, Sean Nassoiy, Paul Shin, Richard Essner. Interpretation of the results: Arthur W. Cowman, Kristel Lourdault, Douglas Hanes, Sean Nassoiy, Paul Shin, Richard Essner. Drafting of the manuscript: Arthur W. Cowman, Kristel Lourdault, Douglas Hanes, Paul Shin, Richard Essner. Critically reviewing or revising the manuscript for important intellectual content: Arthur W. Cowman, Kristel Lourdault, Douglas Hanes, Sean Nassoiy, Paul Shin, Tyler Aguilar, Melanie Goldfarb, Richard Essner. All authors read and approved the final manuscript.
Funding
This study was supported by the Borstein Family Melanoma Program, the Melamed Family Foundation, the John Wayne Cancer Foundation, and the Donald L. Morton Melanoma Research Fund. The Rapid Service Fee was funded by lab grants.
Data Availability
The dataset generated for this study is available upon request to the corresponding author.
Declarations
Conflict of Interest
Richard Essner serves on the advisory board for Castle Biosciences and IntraMedical Imaging. Arthur W. Cowman, Kristel Lourdault, Douglas Hanes, Sean Nassoiy, Paul Shin, Tyler Aguilar, and Melanie Goldfarb have no conflict of interest to report.
Ethical Approval
This study was deemed exempt by the guidelines set forth by the institutional review board. This study did not require ethical approval from an institutional review board (IRB) as it solely involved publicly available data with no identifiable personal information.
References
- 1.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):457–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Duarte-Bateman D, Shen A, Bullock T, et al. Best practices in surgical and nonsurgical management of head and neck Merkel cell carcinoma: an update. Mol Carcinog. 2023;62(1):101–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Siqueira SOM, Campos-do-Carmo G, Dos Santos ALS, Martins C, de Melo AC. Merkel cell carcinoma: epidemiology, clinical features, diagnosis and treatment of a rare disease. An Bras Dermatol. 2023;98(3):277–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Dennis LK, Brown HE, Arrington AK. Comparison of prognostic factors for merkel cell carcinoma, mucosal melanoma and cutaneous malignant melanoma: insights into their etiologies. Curr Oncol. 2023;30(4):3974–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wang AJ, McCann B, Soon WCL, et al. Merkel cell carcinoma: a forty-year experience at the Peter MacCallum Cancer Centre. BMC Cancer. 2023;23(1):30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Becker JC, Stang A, DeCaprio JA, et al. Merkel cell carcinoma. Nat Rev Dis Primers. 2017;3:17077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cass S, Cope B, Bishop AJ, et al. Primary tumor site for localized Merkel cell carcinoma drives different management strategies without impacting oncologic outcomes. Radiother Oncol. 2023;188: 109892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gauci ML, Aristei C, Becker JC, et al. Diagnosis and treatment of Merkel cell carcinoma: European consensus-based interdisciplinary guideline—Update 2022. Eur J Cancer. 2022;171:203–31. [DOI] [PubMed] [Google Scholar]
- 9.Tetzlaff MT, Nagarajan P. Update on Merkel cell carcinoma. Head Neck Pathol. 2018;12(1):31–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Andruska N, Fischer-Valuck BW, Mahapatra L, et al. Association between surgical margins larger than 1 cm and overall survival in patients with Merkel cell carcinoma. JAMA Dermatol. 2021;157(5):540–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Perez MC, de Pinho FR, Holstein A, et al. Resection margins in Merkel cell carcinoma: is a 1-cm margin wide enough? Ann Surg Oncol. 2018;25(11):3334–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lango M, Shnayder Y. Surgical management of Merkel cell carcinoma. Otolaryngol Clin North Am. 2021;54(2):357–68. [DOI] [PubMed] [Google Scholar]
- 13.Gunaratne DA, Howle JR, Veness MJ. Definitive radiotherapy for Merkel cell carcinoma confers clinically meaningful in-field locoregional control: a review and analysis of the literature. J Am Acad Dermatol. 2017;77(1):142–8. [DOI] [PubMed] [Google Scholar]
- 14.Ishizuki S, Nakamura Y. Evidence from clinical studies related to dermatologic surgeries for skin cancer. Cancers. 2022;14:15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jaouen F, Kervarrec T, Caille A, et al. Narrow resection margins are not associated with mortality or recurrence in patients with Merkel cell carcinoma: a retrospective study. J Am Acad Dermatol. 2021;84(4):921–9. [DOI] [PubMed] [Google Scholar]
- 16.Cheraghlou S, Doudican NA, Criscito MC, Stevenson ML, Carucci JA. Overall survival after mohs surgery for early-stage Merkel cell carcinoma. JAMA Dermatol. 2023;159(10):1068–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Singh B, Qureshi MM, Truong MT, Sahni D. Optimal surgical modality for early Merkel cell carcinoma: results from the National Cancer Database. J Am Acad Dermatol. 2023;88(3):e129–30. [DOI] [PubMed] [Google Scholar]
- 18.Carrasquillo OY, Cancel-Artau KJ, Ramos-Rodriguez AJ, Cruzval-O’Reilly E, Merritt BG. Mohs micrographic surgery versus wide local excision in the treatment of merkel cell carcinoma: a systematic review. Dermatol Surg. 2022;48(2):176–80. [DOI] [PubMed] [Google Scholar]
- 19.Jager KJ, van Dijk PC, Zoccali C, Dekker FW. The analysis of survival data: the Kaplan-Meier method. Kidney Int. 2008;74(5):560–5. 10.1038/ki.2008.217. [DOI] [PubMed] [Google Scholar]
- 20.Zabor EC, Assel M. On the Need for Landmark Analysis or Time-dependent Covariates. J Urol. 2023;209(6):1060–62. 10.1097/JU.0000000000003459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yan L, Sun L, Guan Z, Wei S, Wang Y, Li P. Analysis of cutaneous Merkel cell carcinoma outcomes after different surgical interventions. J Am Acad Dermatol. 2020;82(6):1422–34. [DOI] [PubMed] [Google Scholar]
- 22.Tarabadkar ES, Fu T, Lachance K, et al. Narrow excision margins are appropriate for Merkel cell carcinoma when combined with adjuvant radiation: analysis of 188 cases of localized disease and proposed management algorithm. J Am Acad Dermatol. 2021;84(2):340–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Desai AD, Behbahani S, Samie FH. Predictors of time to definitive surgery and survival in Merkel cell carcinoma: analysis of the US National Cancer Database. Clin Exp Dermatol. 2022;47(7):1275–82. [DOI] [PubMed] [Google Scholar]
- 24.Haymerle G, Fochtmann A, Kunstfeld R, Pammer J, Erovic BM. Merkel cell carcinoma: overall survival after open biopsy versus wide local excision. Head Neck. 2016;38(Suppl 1):E1014-1018. [DOI] [PubMed] [Google Scholar]
- 25.Moore KJ, Thakuria M, Ruiz ES. No difference in survival for primary cutaneous Merkel cell carcinoma after Mohs micrographic surgery and wide local excision. J Am Acad Dermatol. 2023;89(2):254–60. [DOI] [PubMed] [Google Scholar]
- 26.Zaidi AJ, Choi J, Nardone B, Liszewski W. Differences in Merkel cell carcinoma survival by site of primary presentation. J Am Acad Dermatol. 2022;86(5):1120–3. [DOI] [PubMed] [Google Scholar]
- 27.Bhatia S, Storer BE, Iyer JG, et al. Adjuvant radiation therapy and chemotherapy in Merkel cell carcinoma: survival analyses of 6908 cases from the National Cancer Data Base. J Natl Cancer Inst. 2016;108:9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The dataset generated for this study is available upon request to the corresponding author.