Evolution of Excisional Surgery Practices for Melanoma in the United States

Key Points

Questions  What is the current landscape of excisional surgery practices for melanoma and how has it changed over time?

Findings  In this cross-sectional study, 79 108 patients with melanomas treated with surgical excision were identified. Marked geographic variation in surgical practices was noted, but the use of Mohs surgery with and without immunohistochemistry is increasing.

Meaning  Excisional practices for melanoma appear to be evolving despite static National Comprehensive Cancer Network guidelines; this shift in surgical practice (ie, the increasing use of Mohs surgery) requires more rigorous investigation in the form of clinical trials to ensure patients are receiving the safest and most efficacious treatment available.

Importance  National Comprehensive Cancer Network guidelines for melanoma have consistently recommended wide local excision as the standard of care since their inception. Although surgery with more comprehensive margin assessment (eg, Mohs surgery) has been advocated for certain subsets of melanoma, how often these techniques are used in clinical practice is uncertain.

Objective  To examine trends in the use of comprehensive margin assessment surgery for melanoma by tracking claims data for Mohs surgery.

Design, Setting, and Participants  This national cross-sectional analysis examined claims data from the Optum Clinformatics Data Mart, a nationally representative database. The study cohort consisted of 79 108 patients undergoing surgical excision for melanoma from January 1, 2001, through December 31, 2016. Data were analyzed from January 1, 2001, through December 31, 2016.

Main Outcomes and Measures  The primary outcome was the likelihood of a melanoma being treated with Mohs surgery over time, evaluated by multivariable logistic regression and expressed as the odds of treatment per additional calendar year.

Results  Among 79 108 patients with melanoma (median age, 63 years [interquartile range {IQR}, 51-73]; 47 407 men [59.9%]), 75 047 were treated with conventional excision (median age, 62 years [IQR, 50-73 years]; 44 786 men [59.7%]) and 4061 with Mohs surgery (median age, 67 years [IQR, 56-76 years]; 2621 men [64.5%]). Mohs surgery was used in 5.1% of all surgical cases, with the rate of Mohs surgery increasing 304% from 2.6% in 2001 to 7.9% in 2016. Odds of receiving Mohs surgery for melanoma increased significantly in more recent calendar years (odds ratio [OR], 1.02 per calendar year; 95% CI, 1.01-1.03; P < .001). Immunohistochemistry (IHC) use was only coded with Mohs surgery in 1087 cases (26.8%), and the odds of receiving Mohs surgery with IHC increased in more recent calendar years (OR, 1.13 per calendar year; 95% CI, 1.10-1.15; P < .001). Use of Mohs surgery and Mohs surgery with IHC for melanoma differed widely across geographic census divisions with greater than 3-fold variation between the regions with highest and lowest use in every period (eg, for 2013 through 2016, the East South Central region used Mohs surgery in 8.8% of melanoma excisions compared with 2.6 in the New England region).

Conclusions and Relevance  Despite stable guidelines for melanoma surgery, the results of this study suggest that surgical practices for melanoma are evolving. Wide variations in surgical practice patterns for melanoma are present in the United States. This study’s findings suggest that the effect of variations in surgical techniques on outcomes requires scrutiny and further study.

National Comprehensive Cancer Network (NCCN) guidelines for the surgical management of melanoma have remained largely static since their inception.¹ Most of the randomized clinical trials that inform clinical practice guidelines were completed more than 10 years ago.² The study design of these clinical trials wisely recognized the difficulty of conforming to standard excision margins in certain anatomical locations. Only 1 randomized clinical trial for invasive melanoma included melanoma on the head and neck,³ and none included melanoma on the hands and feet. The anatomical location of a melanoma influences the likelihood of deviation from the surgical margin recommendations enshrined in melanoma clinical practice guidelines,^4,5 and NCCN guidelines explicitly permit modification of margins “to accommodate individual anatomic or functional considerations”^1(p42) despite the unproven safety of this practice. Adverse outcomes, namely incomplete excision and local recurrence, may be more common for melanomas treated with conventional wide local excision in anatomically constrained and functionally important sites.⁶

Concerns about outcomes for melanomas in cosmetic or functionally important locations have led several surgical groups to selectively use alternative surgical strategies with more comprehensive margin assessment before reconstruction.^7-12 Comprehensive margin assessment surgery often uses en face tissue processing to visualize a greater percentage of the surgical margins¹³ and may be performed with paraffin-embedded permanent sections^10,11 or with frozen sections,^7,9,12 as in Mohs surgery. Mohs surgery was used for 3.5% of all melanomas from 2003 and 2009 according to Surveillance, Epidemiology, and End Results (SEER) data. However, trends outside the 17 SEER registry locations are poorly understood, and marked heterogeneity in use of Mohs surgery between the 17 different SEER registries and declining utilization in 2009 were noted.¹⁴ Updated and more broadly generalizable data are needed to better elucidate how the use of comprehensive margin assessment surgery fits into the current practice landscape for melanoma. In addition, although most of the published research on Mohs for melanoma has used immunohistochemistry (IHC) to aid in visualizing melanocytes in frozen sections,^7,9,12 the penetrance of this practice into the surgical management of melanoma with Mohs surgery is entirely unknown to date.

This study evaluated trends in comprehensive margin assessment surgery for melanoma from 2001 to 2016 via a nationally representative claims database by analyzing the likelihood of using Mohs surgery to treat a melanoma over time. We also evaluated how Mohs surgery for melanoma has evolved over time by analyzing the likelihood of the use of IHC during Mohs surgery over time. Finally, we explored other factors associated with variation in the use of Mohs surgery for melanoma and IHC with Mohs surgery. Understanding how trends in melanoma surgery are evolving in the United States helps ensure that future guidelines are relevant to current practice patterns and may inform future research priorities for melanoma surgery.

This study was a cross-sectional analysis using the Optum Clinformatics Data Mart from January 1, 2001, through December 31, 2016. The Optum database contains deidentified paid claims data for approximately 12 to 14 million annually insured patients in the United States. The Optum patient population has been demonstrated to be similar to the US population in respect to sex, age, and geographic distribution.¹⁵ The data include medical claims along with patient demographic information such as age, sex, educational attainment, income, and geographic location. This study was approved by the institutional review board of the University of Pennsylvania, which waived the need for informed consent for the use of deidentified data. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Patient claims with a diagnosis of primary cutaneous melanoma were identified using International Classification of Diseases, 9th Revision (ICD-9) and International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10). Claims with ICD codes beginning with 172.x (ICD-9) or D03.xx and C43.xx (ICD-10) were queried from the database and were further linked to Current Procedural Terminology codes for surgical excision and IHC staining (eTable 1 in the Supplement). To avoid including repeated claims for the same melanoma treated with staged excision and permanent sections¹⁰ or with surgical excision followed by subsequent reconstruction on a different date, multiple claims for surgical excision of melanoma in the same anatomical location for a patient within 14 days were counted only once. Cases were also excluded when multiple surgical excision codes were used for multiple tumors during a single encounter and the association between excision codes and tumors could not be definitively determined. For each surgically treated melanoma, the following corresponding data points were collected: age, sex, educational level (high school or less, less than bachelor’s degree, or bachelor’s degree or above), income (<$40 000, $40 000-$49 000, $50 000-$59 000, $60 000-$74 000, $75 000-$99 000, or ≥$100 000), census division, anatomical location, and year of surgical treatment. Census divisions included the East North Central (Illinois, Indiana, Michigan, Ohio, and Wisconsin), East South Central (Alabama, Kentucky, Mississippi, and Tennessee), Middle Atlantic (New Jersey, New York, and Pennsylvania), Mountain (Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah, and Wyoming), New England (Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont), Pacific (Alaska, California, Hawaii, Oregon, and Washington), South Atlantic (Delaware, Washington, DC, Florida, Georgia, Maryland, North Carolina, South Carolina, Virginia, and West Virginia), West North Central (Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, and South Dakota), and West South Central (Arkansas, Louisiana, Oklahoma, and Texas) regions of the United States.

Data were analyzed from January 1, 2001, through December 31, 2016. The primary outcome of the likelihood of receiving Mohs surgery to treat a melanoma over time was analyzed descriptively as the relative frequency of Mohs surgery for excisional treatment of melanoma by calendar year, statistically evaluated by multivariable logistic regression, and expressed as the odds of treatment per additional calendar year. Secondary outcomes assessed were the odds of receiving IHC during Mohs surgery per calendar year and factors associated with receiving Mohs surgery for melanoma with and without IHC (including geographic location and sociodemographic factors). Two multivariable logistic regression models were constructed to evaluate associations among several covariates (age, sex, educational attainment, income, census division, anatomical location, and additional calendar year of surgical treatment) and the dependent variables, including the use of Mohs surgery to excise a melanoma (yes or no) and the use of Mohs surgery and IHC (yes or no). All data manipulation and statistical analyses were performed using Stata, version 15 (StataCorp, Inc). Two-sided P < .05 indicated significance.

A total of 79 108 patients were identified with surgical excision of melanomas from 2001 to 2016. Table 1 shows the patient and treatment characteristics of this study population. The median age of the cohort was 63 years (interquartile range [IQR], 51-73 years), with men representing most of the patients (47 407 [59.9%] compared with 31 687 women [40.1%]). Within the study population, 4061 patients (5.1%) were treated with Mohs surgery (median age, 67 years [IQR, 56-76 years]; 2621 men [64.5%] and 1439 women [35.4%]) and 75 047 were treated with conventional excision (median age, 62 years [IQR, 50-73 years]; 44 786 men [59.7%] and 30 248 women [40.3%]).

Figure 1 demonstrates the trend in use of Mohs surgery in the treatment of melanomas, which increased from 2.6% in 2001 to 7.9% in 2016; these data were juxtaposed with data from the SEER registries from 2003 through 2009. The association between the year of melanoma surgery and other covariates is shown in Table 2. The odds of using Mohs surgery for melanoma increased by 1.02 times per calendar year (95% CI, 1.01-1.03; P < .001). The geographic census division of the patient was significantly associated with the odds of having Mohs surgery overall; for example, in 2013 through 2016, the East South Central region used Mohs surgery in 8.8% of melanoma excisions compared with 2.6% of melanoma excisions in the New England region (P < .001). Patients with head and neck melanomas (OR, 2.51; 95% CI, 2.32-2.71; P < .001) and female patients (OR, 1.12; 95% CI, 1.04-1.22; P = .004) had higher odds of receiving Mohs surgery for melanoma (Table 2). Overall, older patients (OR for ≥71 years, 1.71; 95% CI, 1.51-1.94; P < .001), patients with higher household income (OR for≥$100 000 per year, 1.33; 95% CI, 1.17-1.51; P < .001), and patients with higher educational attainment (OR for bachelor’s degree or above, 1.30; 95% CI, 1.14-1.48; P < .001) were associated with higher odds of receiving Mohs surgery for melanoma (Table 2).

Among the melanomas treated with Mohs surgery, the eFigure in the Supplement shows the trend in use of IHC staining. Use of IHC associated with Mohs surgery increased 442% from 2001 to 2016 (from 8.8% in 2001 to 38.9% in 2016). Table 3 shows the patient and treatment characteristics of patients receiving Mohs surgery with and without IHC. The odds of receiving Mohs surgery in conjunction with IHC increased by 1.13 per year (95% CI, 1.10-1.15; P < .001). The geographic census division of the patient was significantly associated with the odds of having Mohs surgery with IHC overall; for example, in 2013 through 2016, the East South Central region used Mohs surgery with IHC in 15.2% of melanoma excisions compared with 51.4% in the West South Central region (P < .001). Patients with head and neck melanomas (OR, 2.64; 95% CI, 2.29-3.05; P < .001) and female patients (OR, 1.17; 95% CI, 1.01-1.36; P = .03) had higher odds of receiving Mohs surgery with IHC for melanoma (Table 2). Overall, patients with higher household income (OR for ≥$100 000 per year, 1.51; 95% CI, 1.18-1.93; P < .001) and patients with higher educational attainment (OR for bachelor’s degree or above, 1.32; 95% CI, 1.04-1.69; P = .02) were more likely to have their melanoma treated with Mohs surgery and IHC (Table 2).

Figure 2 shows a heat map demonstrating variation in the use of Mohs surgery alone and with IHC for melanoma among the geographic census divisions. eTables 2 and 3 in the Supplement provide tabular data showing the sociodemographic and surgical treatment characteristics of each geographic census division. Mohs surgery for melanoma had a greater than 3-fold variation between the lowest (1.8% in New England) and highest (6.8% in Mountain) use within census divisions during the 2001-2016 time frame. Similarly, IHC use with Mohs surgery also demonstrated a greater than 3-fold variation between the lowest (15.4% in East North Central) and highest (50.8% in Mountain) use within census divisions from 2001 to 2016. Greater than 3-fold variation in the use of Mohs surgery and IHC use with Mohs surgery for melanoma was noted in every period. When comparing data from 2001 to 2004 with data from 2014 to 2016, use of Mohs surgery for melanoma increased in all but 1 (East North Central) geographic census division, and the use of IHC with Mohs surgery for melanoma increased in all geographic census divisions.

The surgical treatment of melanoma in the United States is evolving in the face of static NCCN guidelines and no new randomized clinical trial data. Mohs surgery is increasingly used to excise melanomas in the United States, with a 2.6% rate in 2001 and 7.9% in 2016 (Figure 1). When Mohs surgery is used, IHC staining has become more common, although it remains used in less than 40% of Mohs surgery cases. Greater than 3-fold variation in the use of Mohs surgery for melanomas and in IHC use with Mohs surgery exists across geographic census divisions in the United States.

The likelihood of a melanoma being treated with Mohs surgery has more than tripled from 2001 to 2016 in the United States. These findings update and add to the limited information previously available about the use of Mohs surgery for melanoma in the United States.¹⁴ Given that Mohs surgery is being used in a growing fraction of melanomas, there is increased urgency to study the safety of this shift in melanoma care. Although several recent database studies have suggested that Mohs surgery appears to have mortality outcomes similar to those for conventional excision,^16-18 population-based observational study results in oncology often do not mirror the outcomes derived from clinical trials.¹⁹ In addition, although the risk of death may be similar between surgical techniques, the more salient question is whether Mohs surgery actually delivers lower rates of local recurrence when compared directly with conventional excision; this benefit is necessary to justify the procedural complexity of incorporating Mohs surgery or other comprehensive margin assessment surgery techniques into widespread practice.

Coinciding with the increase in use of Mohs surgery has been an increase in the use of IHC with Mohs surgery for melanoma. The first cohort study of melanoma treated exclusively with Mohs surgery supplemented with IHC was published in 2000,²⁰ and IHC has been advocated as a method to overcome the challenges of visualizing melanocytes on frozen sections.^21,22 Frozen sections with IHC are similar in ease of interpretation to permanent sections with IHC,²² and historically low rates of local recurrence have uniformly been observed for Mohs surgery aided by immunostaining for melanoma.^7,9,12 In contrast, Mohs surgery without IHC has been associated with as much as a 33% local recurrence rate²³; concerns about the widespread use of Mohs surgery are legitimate when important variations in practice exist.

Unwarranted variations in care can lead to unnecessary variations in outcome.²⁴ Variation in the use of Mohs surgery and the use of IHC with Mohs surgery is widespread and persistent in the United States. Greater than 3-fold variations in the rate of using Mohs surgery for melanoma and the use of IHC with Mohs surgery were observed between the regions with the highest and lowest use in all years (eTable 3 in the Supplement). Variations in surgical care of cancer may be more common when specialty society and NCCN guidelines are discordant.²⁵ Disparate recommendations about when comprehensive margin assessment surgery for melanoma is appropriate^1,26,27 may contribute to wide variation in practice patterns across the United States.

In addition to geography, several patient and melanoma-specific factors were associated with the use of Mohs surgery for melanoma. Anatomical location on the head and neck was associated with increased odds of receiving Mohs surgery for melanoma because concerns about the practical challenges of recommended wide local excision margins^1,4,5 and about higher rates of adverse surgical outcomes with conventional excision on anatomically and functionally constrained locations⁶ are often the stated rationale for using comprehensive margin assessment surgery in select cases. Explanations for the association between several sociodemographic factors (increased age, female sex, household income, and educational attainment) and the use of Mohs surgery for melanoma require further study. However, the use of comprehensive margin assessment surgery seems to be more prevalent for persons with greater income and educational attainment, suggesting barriers to access that may be socioeconomic in nature.

Although these data provide a longitudinal view of changes in excisional surgery for melanoma in the United States that are based on a demographically representative cohort, they have inherent limitations. Trends outside the United States remain uncertain. In addition, comprehensive margin assessment surgery performed with permanent sections is typically billed with excision codes for malignant neoplasms, so our data likely underrepresent the overall use of margin-controlled surgery for melanoma. Finally, although claims-based analyses may be subject to misclassification errors due to incorrect billing practices, changing trends in misclassification of billing practices are unlikely to account for the trends of excisional surgery for melanoma seen in this data.

Despite stable guidelines and no randomized clinical trial data evaluating outcomes of comprehensive margin assessment surgery for melanoma, surgical excision practices for melanoma continue to evolve. The use of Mohs surgery for melanoma has more than tripled and is accompanied by wide variations in practice patterns across the United States. Given the increasing use of Mohs surgery to surgically excise melanomas, there is greater urgency to evaluate the safety and efficacy of comprehensive margin assessment surgery compared with wide local excision with randomized prospective trials. High-quality evidence evaluating different methods of excision (as opposed to different excisional margins) would better inform consensus guideline statements about the role of comprehensive margin assessment surgery for melanoma and possibly reduce variations in practice patterns and care.

Accepted for Publication: June 19, 2019.

Corresponding Author: Jeremy R. Etzkorn, MD, Department of Dermatology, Hospital of the University of Pennsylvania, 3400 Civic Center Blvd, Ste 1-330S, Philadelphia, PA 19104 (jeremy.etzkorn@pennmedicine.upenn.edu).

Published Online: August 28, 2019. doi:10.1001/jamadermatol.2019.2346

Author Contributions: Drs Etzkorn and Lee had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Lee, Barbieri, Miller, Etzkorn.

Acquisition, analysis, or interpretation of data: Lee, Sobanko, Shin, Howe, Barbieri, Etzkorn.

Drafting of the manuscript: Lee, Miller, Etzkorn.

Critical revision of the manuscript for important intellectual content: Sobanko, Shin, Howe, Barbieri, Miller, Etzkorn.

Statistical analysis: Lee, Etzkorn.

Obtained funding: Lee, Etzkorn.

Administrative, technical, or material support: Lee, Howe, Barbieri.

Supervision: Etzkorn.

Conflict of Interest Disclosures: Dr Lee reported receiving grants from the Melanoma Research Foundation during the conduct of the study. Dr Shin reported receiving grants from Regeneron Pharmaceuticals, Inc, outside the submitted work. Dr Etzkorn reported receiving grants from the Melanoma Research Foundation during the conduct of the study. No other disclosures were reported.

Funding/Support: This study was supported by a Medical Student Research Grant from the Melanoma Research Foundation (Mr Lee); by award T32-AR-007465 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health (Dr Barbieri); by partial salary support through a Fellowship in Dermatology Patient Oriented Research grant to the Trustees of the University of Pennsylvania from Pfizer, Inc (Dr Barbieri); and by a Career Development Award in Dermatologic Surgery from the Dermatology Foundation (Dr Etzkorn).

Role of the Funder/Sponsor: The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: Dr Etzkorn is an associate editor of JAMA Dermatology, but he was not involved in any of the decisions regarding review of the manuscript or its acceptance.