- Email: [email protected]
Application of Mohs micrographic surgery appropriate-use criteria to skin cancers at a university health system
ORIGINAL
ARTICLE
Application of Mohs micrographic surgery appropriate-use criteria to skin cancers at a University Health System Adam B. Blechman, MD,a James W. Patterson, MD,a,b and Mark A. Russell, MDa Charlottesville, Virginia Background: Mohs micrographic surgery (MMS) is an effective treatment for skin cancer. Until recently, it has been difficult to determine the percentage of skin cancers best treated with MMS. The appropriate-use criteria represents an opportunity to more accurately estimate this number. Objective: We sought to apply the appropriate-use criteria retrospectively to University of Virginia Health System skin cancers so as to determine the proportion that met appropriate use within 8 months. Methods: A list of all biopsy-proven skin cancers, excluding invasive melanoma, at the University of Virginia Health System during an 8-month period was generated. Patient and tumor data were collected retrospectively from hospital records and each skin cancer was classified as appropriate, inappropriate, or uncertain based on the appropriate-use criteria. Results: Among 1059 skin cancers, MMS was appropriate in 72.0% of cases, inappropriate in 20.4%, and uncertain in 7.6%. Altogether, 59.3% of skin cancers occurred in H and M areas, which include the head, neck, hands, feet, ankles, genitalia, nipples/areola, and pretibial surface. Limitations: Patient and tumor information was collected retrospectively at 1 institution. Conclusions: Using recently published appropriate-use criteria, 72.0% of skin cancers at this institution were appropriate for MMS. Tumor location was the most important factor in determining appropriate use. ( J Am Acad Dermatol http://dx.doi.org/10.1016/j.jaad.2014.02.025.) Key words: appropriate-use criteria; Mohs micrographic surgery; skin cancer.
S
kin cancer is the most common form of cancer in the United States, and the incidence continues to rise. It is estimated that there were 3.5 million annual nonmelanoma skin cancers1 in 2006 and that 1 of 5 people in the United States will get skin cancer in their lifetime.2 Likewise, the incidence of melanoma in situ (MIS) has increased 62% over the last 10 years from 34,300 in 2002 to 55,560 in 2012.3,4 Fortunately, there are many effective treatments for skin cancer including Mohs micrographic surgery (MMS). The MMS appropriate-use criteria was created through a joint effort by the American Academy of From the Departments of Dermatologya and Pathology,b University of Virginia Health System. Funding sources: None. Conflicts of interest: None declared. This project, including the diagram of H, M, and L areas, was presented at the Annual Meeting of the American College of Mohs Surgery, Washington, District of Columbia, May 4, 2013. Accepted for publication February 18, 2014.
Abbreviations used: BCC: MIS: MMS: SCC: SCCIS:
basal cell carcinoma melanoma in situ Mohs micrographic surgery squamous cell carcinoma squamous cell carcinoma in situ
Dermatology, American College of Mohs Surgery, American Society for Mohs Surgery, and American Society for Dermatologic Surgery. The appropriate-use criteria task force, composed of more than 70 dermatologists, outlined appropriate-use criteria for 270
Reprint requests: Mark A. Russell, MD, Department of Dermatology, University of Virginia Health System, Box 800718, Charlottesville, VA 22908. E-mail: [email protected]. Published online May 07, 2014. 0190-9622/$36.00 Ó 2014 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2014.02.025
1
2 Blechman, Patterson, and Russell
J AM ACAD DERMATOL
practices but diagnosed by a University of clinical situations of skin cancer based on characterisVirginia dermatopathologist were not included tics such as tumor location, size, recurrence, and because patient records were not readily availhistology. A 17-member ratings panel consisting of able. Even though a study showed that rare skin Mohs surgeons and non-Mohs dermatologists cancers typically represent less than 1% of all labeled each clinical situation into appropriate, inaptumors treated by Mohs surgeons, they were propriate, or uncertain categories. Among 270 included in this study for completeness.7 situations, 200 (74.1%) were deemed as appropriate, 24 (8.9%) as inappropriate, Invasive melanomas were and 46 (17.0%) as uncertain. excluded, as these were CAPSULE SUMMARY In October 2012, the not addressed in the appropriate-use criteria were appropriate-use criteria. Appropriate-use criteria for Mohs dually published in both the micrographic surgery were recently Journal of the American Histology published. 5 Academy of Dermatology Biopsies were performed This article reports the percent of skin and Dermatologic Surgery.6 using shave, punch, and incicancers that met appropriate-use criteria Until recently, it was sional techniques. Histologic at a University Health System. difficult to estimate the perinformation was collected centage of skin cancers from pathology reports. Although most skin cancers biopsied in appropriate for MMS because Basal cell carcinomas (BCCs) this study met appropriate-use criteria, there was no universally were grouped into nodular, the majority were not treated by this accepted standard of practice superficial, and aggressive modality. within the field. Yet, the subtypes as described by appropriate-use criteria repthe appropriate-use criteria. resents a method to estimate this number. This The appropriate-use criteria uses several aggressive retrospective study will apply the appropriate-use features including morpheaform/fibrosing/sclecriteria to past University of Virginia Health rosing, infiltrating, perineural, metatypical/keratotic, System skin cancers to determine the proportion and micronodular. Similarly, the appropriate-use that met appropriate use within an 8-month time criteria identifies aggressive features for squamous period. cell carcinomas (SCCs). These types include sclerosing, basosquamous, poorly or undifferentiated, perineural/perivascular, infiltrating, keratoacanthoma (central facial), clear cell, sarcomatoid, METHODS Breslow depth 2 mm or more, Clark level 4 or Patient selection greater, pagetoid, single cell, lymphoepithelial, and A list of all skin biopsyeproven skin cancers at small cell. Subtypes not mentioned by the the University of Virginia Health System from May appropriate-use criteria were grouped under one 2011 to the end of December 2011 was generated of the categories after review of the slides by a using the CoPath database system. Skin biopsies University of Virginia dermatopathologist. If multiperformed by University of Virginia health prople subtypes existed in 1 specimen, the most fessionals and diagnosed by University of Virginia aggressive feature was used. dermatopathologists were included in this study. Whether a dermatologist or nondermatologist Skin cancer characteristics (location, size, physician performed the biopsy was noted. For recurrence) patients who underwent multiple skin biopsies, The appropriate-use criteria delineates 3 areas each lesion was recorded as an independent data of the body (H, M, L) and 4 increments of size point. ( # 0.5, 0.6-1.0, 1.1-2.0, and [2 cm). Skin cancer Patient information was collected retrospeclocation and size was determined using descriptively from hospital records with institutional retions from patient notes and pathology reports. view board approval. Skin cancer characteristics Figs 1 and 2 illustrate areas designated as H, M, such as location, pretreatment size, histologic type and L based on the appropriate-use criteria and subtype, recurrence, and first treatment type outline: were recorded for each lesion. Also, patient Area H: ‘‘mask’’ areas of face (central aspect of face, demographics including age, gender, history of eyelids [including inner/outer canthi], eyebrows, immunosuppression, history of radiation therapy, nose, lips [cutaneous/mucosal/vermillion], chin, and history of transplantation procedures were ear and periauricular skin/sulci, temple), genitalia tabulated. Skin biopsies performed at external d
d
d
J AM ACAD DERMATOL
Blechman, Patterson, and Russell 3
Fig 2. Body H, M, and L areas. Fig 1. Head and neck H and M areas.
(including perineal and perianal), hands, feet, nail units, ankles, and nipples/areola. Area M: cheeks, forehead, scalp, neck, jawline, and pretibial surface. Area L: trunk and extremities (excluding pretibial surface, hands, feet, nail units, and ankles).5,6 Skin cancers were deemed recurrent if patient notes documented a prior lesion treated in the same location or if the lesion was arising out of a previous treatment scar. Otherwise, skin cancers were considered primary. Patient characteristics (immunosuppression, transplantation, radiation, syndromes) The appropriate-use criteria note several patient characteristics including immunosuppression and transplantation status, previous irradiation to the skin cancer area, and history of genetic syndromes. These characteristics were documented using patient records. A patient was listed as immunosuppressed if they had a history of HIV, organ transplantation, hematologic malignancy, or pharmacologic immunosuppression. Treatment For skin biopsies occurring in the study period, patient records were followed up to the point of
treatment even if it was outside the study period. If a patient received multiple treatments, the initial treatment was counted. Treatment types included topical therapy (imiquimod, 5-fluorouracil), cryotherapy, curettage and electrodessication, excision, MMS, and radiation. MMS cases were referred to a University of Virginia Mohs surgeon. MIS/lentigo malignas were referred to a surgical oncologist, plastic surgeon, or otorhinolaryngologist for excision. Appropriate-use criteria After all patient and tumor characteristics were collected, each skin cancer was classified as appropriate, inappropriate, or uncertain based on the appropriate-use criteria. If the biopsy specimen was too superficial to determine histologic information, a classification was attempted using remaining information. For example, all BCCs in H areas are appropriate for MMS, regardless of subtype, based on the appropriate-use criteria. If the skin cancer was too superficial and could not be classified using the remaining information, it was excluded. Data analysis The percentages of skin cancers falling under appropriate, inappropriate, or uncertain designation were calculated. Percentages for patient
J AM ACAD DERMATOL
4 Blechman, Patterson, and Russell
Fig 3. Skin cancer selection. AUC, Appropriate-use criteria.
characteristics, tumor characteristics, and treatment types were determined. Standard statistics software showed that the sample size of this study can estimate the percentage of skin cancers appropriate for MMS for a similar population of any size with a 95% confidence level and a 63.0% confidence interval. A 2-tailed Fisher exact test was used to determine any statistically significant difference in MMS referral patterns between dermatologists and nondermatologist physicians. A P value equal to or less than .05 was deemed statistically significant. The statistical analysis was performed using software (SPSS Statistics 20.0, IBM Corp, Armonk, NY).
RESULTS From May 2011 to December 2011, a total of 1070 BCCs, SCCs, SCCs in situ (SCCISs), MIS/lentigo malignas, and other rare tumors were diagnosed by biopsy at the University of Virginia Health System. Of these, 11 biopsy specimens were too superficial to determine histologic characteristics necessary to make an appropriate-use designation (Fig 3). This left 1059 skin cancers with a breakdown of 667 (63.0%) BCCs, 244 (23.0%) SCCs, 108 (10.2%) SCCISs, 28 (2.6%) MIS/lentigo malignas, and 12 (1.1%) rare tumors. The rare tumors included 5 (0.5%) Merkel cell carcinomas, 2 (0.2%) angiosarcomas, 2 (0.2%) dermatofibrosarcoma protuberans, 1 (0.1%) leiomyosarcoma, 1 (0.1%) extramammary Paget disease, and 1 (0.1%) malignant fibrous histiocytoma. The location of these skin cancers varied among the H, M, and L areas with incidences of 33.4%, 25.9%, and 40.7%, respectively. Most pretreatment skin cancer sizes were in the range of 0.6 to 1.0 cm
(46.4%), followed by 1.1 to 2.0 cm (30.5%), less than or equal to 0.5 cm (16.0%), and greater than 2.0 cm (7.1%). The mean age was 68 years old with a range of 19 to 101 years. The majority (64.6%) of skin cancers were diagnosed in men. In 116 or 11.0% of all skin cancers, the patients were immunosuppressed. Of the 667 BCCs, 27.0% were aggressive subtypes, 21.4% nodular subtypes, and 30.1% superficial subtypes. Overall, 72.0% of skin cancers were appropriate for MMS based on the appropriate-use criteria. Likewise, 20.4% were deemed inappropriate and 7.6% uncertain. Among skin cancer types, SCCs had the highest proportion of skin cancers falling in the appropriate category at 90.6% (5.7% inappropriate, 3.7% uncertain). In the BCC group, 67.3%, 27.6%, and 5.1% were appropriate, inappropriate, and uncertain, respectively. In the SCCIS group, 63.0%, 16.7%, and 20.4% were appropriate, inappropriate, and uncertain, respectively. In the MIS group, 67.9%, 0.0%, and 32.1% were appropriate, inappropriate, and uncertain, respectively. Lastly, among all rare skin cancers, 50.0% were appropriate and 50.0% were uncertain. These percentages are summarized in Table I. The most common treatment types were excision (36.5%), MMS (29.7%), and electrodesiccation and curettage (22.2%). Other treatment statistics can be viewed in Table II. Among 1059 skin cancers, dermatologists diagnosed 987 and nondermatologist physicians identified 72. The proportion of skin cancers meeting appropriate use for MMS was 71.3% for dermatologists and 80.6% for nondermatologists, which was not a statistically significant difference (Table III).
J AM ACAD DERMATOL
Blechman, Patterson, and Russell 5
Table I. Appropriate-use criteria by skin cancer type Appropriate
BCC SCC SCCIS MIS/LM Other Total
449 221 68 19 6 763
(67.3) (90.6) (63.0) (67.9) (50.0) (72.0)
Inappropriate
184 14 18 0 0 216
(27.6) (5.7) (16.7) (0.0) (0.0) (20.4)
Uncertain
34 9 22 9 6 80
(5.1) (3.7) (20.4) (32.1) (50.0) (7.6)
Percentages of each skin cancer type are in parentheses. BCC, Basal cell carcinoma; LM, lentigo maligna; MIS, melanoma in situ; SCC, squamous cell carcinoma; SCCIS, squamous cell carcinoma in situ.
Table II. Treatment types Imiquimod, 5-fluorouracil Cryotherapy Electrodesiccation and curettage Excision Mohs micrographic surgery Radiation Other* Total
37 9 235 386 315 0 77 1059
(3.5) (0.8) (22.2) (36.5) (29.7) (0.0) (7.3) (100.0)
Percentages of all skin cancers are in parentheses. *Includes skin cancers that were completely removed by shave biopsy, that never received definitive treatment after shave biopsy, or for which the outcome was unknown.
More skin cancers appropriate for MMS were subsequently referred for MMS in the dermatologist group compared with the nondermatologist group (43.6% vs 13.8%, respectively). This difference was statistically significant (P = .000).
DISCUSSION The Medicare claims database from 2009 showed a 236% increase in MMS over 10 years. Part of this rise can be attributed to an increase in skin cancer prevalence. However, excisions and destructions increased by only 20% over the same time frame.8 Other reasons for the MMS surge include an increase in Mohs surgeons and a higher awareness by referring providers to the benefits of MMS. Although studies have shown MMS to have lower recurrence rates than other treatments,9 it is more expensive than some other treatment options, requires special training, requires special laboratory facilities, and is not available everywhere. Therefore, referring providers should understand which lesions are appropriate and inappropriate for MMS, and know treatment alternatives. In 1995, the American Academy of Dermatology proposed guidelines for MMS10; however, these lacked the level of detail
Table III. Mohs micrographic surgery referral patterns among dermatologists and nondermatologists Dermatologists
Skin cancers diagnosed Skin cancers meeting appropriate use Skin cancers referred for MMS Proportion of appropriate ultimately referred
Nondermatologists
P*
987 (100.0)
72 (100.0)
704 (71.3)
58 (80.6)
.103
307 (31.1)
8 (11.1)
.000
307/704 (43.6)
8/58 (13.8)
.000
Percentages of total skin cancers in each group are in parentheses. Dermatologists and nondermatologists had similar proportions of skin cancers meeting appropriate use for MMS, but more skin cancers appropriate for MMS were subsequently referred for MMS in the dermatologist group compared with the nondermatologist group. MMS, Mohs micrographic surgery. *Two-tailed Fisher exact test.
seen with the appropriate-use criteria published in 2012. In this study, tumor location was the most common factor used to establish which skin cancers met appropriate use. This finding was a result of 2 reasons. First, a large number of skin cancers occurred in H and M areas. Second, most skin cancers arising in H and M areas are designated appropriate for MMS by the appropriate-use criteria. Among the 72.0% of skin cancers that met appropriate use, 59.3% of them were in H and M areas. Under the appropriate-use criteria, all SCCs, BCCs, SCCISs, and MIS/lentigo malignas, except primary actinic keratosis (AKs) with focal SCCIS, are appropriate if located in the H area. H area skin cancers represented 33.4% of the sample size. Similarly, all skin cancers in the M area are appropriate except for superficial BCCs equal to or less than 5 mm in size and primary AKs with focal SCCIS, which accounted for another 25.8% of all lesions. Aggressive histologic features also contributed to a large proportion of appropriate lesions, especially for SCCs in which 90.6% were appropriate for MMS and 47.5% had aggressive subtypes. According to the appropriate-use criteria, any aggressive histologic feature for SCC makes it appropriate for MMS, regardless of all other characteristics. Among 244 SCCs, the most common aggressive features were Breslow depth 2 mm or more (30.7%), poor
6 Blechman, Patterson, and Russell
differentiation (8.2%), or both (2.0%). One limitation of the appropriate-use criteria is that it does not address all important tumor subtypes. For example, it does not include acantholytic or intravascular tumors as an aggressive subtype of SCC. Likewise, all BCCs are categorized into superficial, nodular, or aggressive subtypes even though there are many others. In these few situations, a University of Virginia dermatopathologist (J. W. P.) classified each skin cancer as best as possible using the information provided by the appropriate-use criteria. Skin cancers appropriate for MMS but not treated with MMS received some other treatment; electrodesiccation with curettage and excision were the most common other treatment types (Table II). The gap between 72% of skin cancers appropriate for MMS and only 29.7% referred for MMS was a result of various reasons such as patient comorbidities, scheduling constraints, travel constraints, and personal preference. Further, MMS was not used for MIS/ lentigo maligna although there were only 28 MIS/ lentigo malignas comprising less than 0.5% of our tumors. Also, skin cancers in this study were diagnosed and treated in 2011, which was before the appropriate-use criteria were published. Nevertheless, it implies that MMS might be underused. These statistics are not just specific to University of Virginia but studies from other institutions have cited similar MMS use rates of 30% to 35%.11-13 Dermatologists had a significantly higher MMS referral rate for appropriate skin cancers than that of nondermatologist physicians. Interestingly, of the 50 skin cancers diagnosed by nondermatologists that met appropriate use but were not referred for MMS, 35 or 70% were appropriate because of their locations in H and M areas. Likewise, in the dermatologist group, 73% of skin cancers meeting appropriate use but ultimately referred for other treatment methods were appropriate because of their location in H and M areas. Therefore, better understanding of H and M areas by both dermatologists and especially nondermatologists may lead to more skin cancers appropriate for MMS to be ultimately referred and treated with MMS. The purpose of this study was to apply the appropriate-use criteria to skin cancers at the University of Virginia Health System to determine the percent meeting appropriate use. These results provide insight into the potential demand for MMS at this institution but also other academic institutions. The percentages can also be applied on a national level to extrapolate the potential demand for MMS in the United States. By applying the 72.0%
J AM ACAD DERMATOL
appropriate-use figure to 2006 estimates of 3.5 million nonmelanoma skin cancers1 in the United States, 2.5 million would have been appropriate for MMS. This study’s sample size is large enough to make estimates of this magnitude within 3.0%. Currently, although most skin cancers may be appropriate for MMS, most were not treated by this modality. The ability to evaluate national trends may be limited because of differences in patient and tumor characteristics between academic and private practice settings. To address this issue, similar studies are warranted that encompass both academic and private practice patients. Also, because this study was retrospective, patient characteristics such as history of treatment and recurrence, radiation treatment to the area, genetic syndromes, and immunosuppression might have not been accounted for in some patients; this would underestimate the percent of cancers appropriate for MMS. Conclusion By applying the recently published appropriateuse criteria to histologically diagnosed skin cancers, it was this institution’s experience that 72.0% were appropriate for MMS. This percent was based not only on histology but also on anatomic location, pretreatment size, prior treatment, prior radiotherapy, and presence of immunologic problems. Tumor location was the most important factor, as almost all skin cancers in H and M areas were appropriate. Dermatologists were more likely than nondermatologist physicians to refer skin cancers that met appropriate use. Widespread understanding of H and M areas by both dermatologists and especially nondermatologists may lead to more skin cancers appropriate for MMS to be ultimately referred and treated with MMS. REFERENCES 1. Rogers HW, Weinstock MA, Harris AR, Hinckley MR, Feldman SR, Fleischer AB, Coldiron BM. Incidence estimate of nonmelanoma skin cancer in the United States, 2006. Arch Dermatol 2010;146:283-7. 2. Stern RS. Prevalence of a history of skin cancer in 2007: results of an incidence-based model. Arch Dermatol 2010;146: 279-82. 3. American Cancer Society. Cancer facts and figures 2012. Available from: URL:http://www.cancer.org/acs/groups/content/ @epidemiologysurveilance/documents/document/acspc-031941. pdf. Accessed December 10, 2012. 4. American Cancer Society. Cancer facts and figures 2002. Available from: URL:http://www.uhmsi.com/docs/CancerFacts& Figures2002.pdf. Accessed December 10, 2012. 5. Ad Hoc Task Force, Connolly SM, Baker DR, Coldiron BM, Fazio MJ, Storrs PA, Vidimos AT, et al. AAD/ACMS/ASDSA/ASMS 2012 appropriate use criteria for Mohs micrographic surgery: a report of the American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic
J AM ACAD DERMATOL
Surgery Association, and the American Society for Mohs Surgery. J Am Acad Dermatol 2012;67:531-50. 6. American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, Connolly SM, Baker DR, Coldiron BM, Fazio MJ, Storrs PA, Vidimos AT, et al. AAD/ACMS/ASDSA/ASMS 2012 appropriate use criteria for Mohs micrographic surgery: a report of the American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, and the American Society for Mohs Surgery. Dermatol Surg 2012;38:1582-603. 7. Anderson HL, Joseph AK. A pilot feasibility study of a rare skin tumor database. Dermatol Surg 2007;33:693-6. 8. Donaldson MR, Coldiron BM. Mohs micrographic surgery utilization in the Medicare population, 2009. Dermatol Surg 2012;38:1427-34.
Blechman, Patterson, and Russell 7
9. Thissen MR, Neumann MH, Schouten LJ. A systematic review of treatment modalities for primary basal cell carcinomas. Arch Dermatol 1999;135:1177-83. 10. Drake LA, Dinehart SM, Goltz RW, Graham GF, Hordinsky MK, Lewis CW, et al. Guidelines of care for Mohs micrographic surgery: American Academy of Dermatology. J Am Acad Dermatol 1995;33:271-8. 11. Donaldson MR, Coldiron BM. Mohs micrographic surgery utilization in the Medicare population, 2009. Dermatol Surg 2012;38:1427-34. 12. Gaston DA, Naugle C, Clark DP. Mohs micrographic surgery referral patterns: the University of Missouri experience. Dermatol Surg 1999;25:862-7. 13. Welch ML, Anderson LL, Grabski WJ. How many nonmelanoma skin cancers require Mohs micrographic surgery? Dermatol Surg 1996;22:711-3.
ARTICLE
Application of Mohs micrographic surgery appropriate-use criteria to skin cancers at a University Health System Adam B. Blechman, MD,a James W. Patterson, MD,a,b and Mark A. Russell, MDa Charlottesville, Virginia Background: Mohs micrographic surgery (MMS) is an effective treatment for skin cancer. Until recently, it has been difficult to determine the percentage of skin cancers best treated with MMS. The appropriate-use criteria represents an opportunity to more accurately estimate this number. Objective: We sought to apply the appropriate-use criteria retrospectively to University of Virginia Health System skin cancers so as to determine the proportion that met appropriate use within 8 months. Methods: A list of all biopsy-proven skin cancers, excluding invasive melanoma, at the University of Virginia Health System during an 8-month period was generated. Patient and tumor data were collected retrospectively from hospital records and each skin cancer was classified as appropriate, inappropriate, or uncertain based on the appropriate-use criteria. Results: Among 1059 skin cancers, MMS was appropriate in 72.0% of cases, inappropriate in 20.4%, and uncertain in 7.6%. Altogether, 59.3% of skin cancers occurred in H and M areas, which include the head, neck, hands, feet, ankles, genitalia, nipples/areola, and pretibial surface. Limitations: Patient and tumor information was collected retrospectively at 1 institution. Conclusions: Using recently published appropriate-use criteria, 72.0% of skin cancers at this institution were appropriate for MMS. Tumor location was the most important factor in determining appropriate use. ( J Am Acad Dermatol http://dx.doi.org/10.1016/j.jaad.2014.02.025.) Key words: appropriate-use criteria; Mohs micrographic surgery; skin cancer.
S
kin cancer is the most common form of cancer in the United States, and the incidence continues to rise. It is estimated that there were 3.5 million annual nonmelanoma skin cancers1 in 2006 and that 1 of 5 people in the United States will get skin cancer in their lifetime.2 Likewise, the incidence of melanoma in situ (MIS) has increased 62% over the last 10 years from 34,300 in 2002 to 55,560 in 2012.3,4 Fortunately, there are many effective treatments for skin cancer including Mohs micrographic surgery (MMS). The MMS appropriate-use criteria was created through a joint effort by the American Academy of From the Departments of Dermatologya and Pathology,b University of Virginia Health System. Funding sources: None. Conflicts of interest: None declared. This project, including the diagram of H, M, and L areas, was presented at the Annual Meeting of the American College of Mohs Surgery, Washington, District of Columbia, May 4, 2013. Accepted for publication February 18, 2014.
Abbreviations used: BCC: MIS: MMS: SCC: SCCIS:
basal cell carcinoma melanoma in situ Mohs micrographic surgery squamous cell carcinoma squamous cell carcinoma in situ
Dermatology, American College of Mohs Surgery, American Society for Mohs Surgery, and American Society for Dermatologic Surgery. The appropriate-use criteria task force, composed of more than 70 dermatologists, outlined appropriate-use criteria for 270
Reprint requests: Mark A. Russell, MD, Department of Dermatology, University of Virginia Health System, Box 800718, Charlottesville, VA 22908. E-mail: [email protected]. Published online May 07, 2014. 0190-9622/$36.00 Ó 2014 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2014.02.025
1
2 Blechman, Patterson, and Russell
J AM ACAD DERMATOL
practices but diagnosed by a University of clinical situations of skin cancer based on characterisVirginia dermatopathologist were not included tics such as tumor location, size, recurrence, and because patient records were not readily availhistology. A 17-member ratings panel consisting of able. Even though a study showed that rare skin Mohs surgeons and non-Mohs dermatologists cancers typically represent less than 1% of all labeled each clinical situation into appropriate, inaptumors treated by Mohs surgeons, they were propriate, or uncertain categories. Among 270 included in this study for completeness.7 situations, 200 (74.1%) were deemed as appropriate, 24 (8.9%) as inappropriate, Invasive melanomas were and 46 (17.0%) as uncertain. excluded, as these were CAPSULE SUMMARY In October 2012, the not addressed in the appropriate-use criteria were appropriate-use criteria. Appropriate-use criteria for Mohs dually published in both the micrographic surgery were recently Journal of the American Histology published. 5 Academy of Dermatology Biopsies were performed This article reports the percent of skin and Dermatologic Surgery.6 using shave, punch, and incicancers that met appropriate-use criteria Until recently, it was sional techniques. Histologic at a University Health System. difficult to estimate the perinformation was collected centage of skin cancers from pathology reports. Although most skin cancers biopsied in appropriate for MMS because Basal cell carcinomas (BCCs) this study met appropriate-use criteria, there was no universally were grouped into nodular, the majority were not treated by this accepted standard of practice superficial, and aggressive modality. within the field. Yet, the subtypes as described by appropriate-use criteria repthe appropriate-use criteria. resents a method to estimate this number. This The appropriate-use criteria uses several aggressive retrospective study will apply the appropriate-use features including morpheaform/fibrosing/sclecriteria to past University of Virginia Health rosing, infiltrating, perineural, metatypical/keratotic, System skin cancers to determine the proportion and micronodular. Similarly, the appropriate-use that met appropriate use within an 8-month time criteria identifies aggressive features for squamous period. cell carcinomas (SCCs). These types include sclerosing, basosquamous, poorly or undifferentiated, perineural/perivascular, infiltrating, keratoacanthoma (central facial), clear cell, sarcomatoid, METHODS Breslow depth 2 mm or more, Clark level 4 or Patient selection greater, pagetoid, single cell, lymphoepithelial, and A list of all skin biopsyeproven skin cancers at small cell. Subtypes not mentioned by the the University of Virginia Health System from May appropriate-use criteria were grouped under one 2011 to the end of December 2011 was generated of the categories after review of the slides by a using the CoPath database system. Skin biopsies University of Virginia dermatopathologist. If multiperformed by University of Virginia health prople subtypes existed in 1 specimen, the most fessionals and diagnosed by University of Virginia aggressive feature was used. dermatopathologists were included in this study. Whether a dermatologist or nondermatologist Skin cancer characteristics (location, size, physician performed the biopsy was noted. For recurrence) patients who underwent multiple skin biopsies, The appropriate-use criteria delineates 3 areas each lesion was recorded as an independent data of the body (H, M, L) and 4 increments of size point. ( # 0.5, 0.6-1.0, 1.1-2.0, and [2 cm). Skin cancer Patient information was collected retrospeclocation and size was determined using descriptively from hospital records with institutional retions from patient notes and pathology reports. view board approval. Skin cancer characteristics Figs 1 and 2 illustrate areas designated as H, M, such as location, pretreatment size, histologic type and L based on the appropriate-use criteria and subtype, recurrence, and first treatment type outline: were recorded for each lesion. Also, patient Area H: ‘‘mask’’ areas of face (central aspect of face, demographics including age, gender, history of eyelids [including inner/outer canthi], eyebrows, immunosuppression, history of radiation therapy, nose, lips [cutaneous/mucosal/vermillion], chin, and history of transplantation procedures were ear and periauricular skin/sulci, temple), genitalia tabulated. Skin biopsies performed at external d
d
d
J AM ACAD DERMATOL
Blechman, Patterson, and Russell 3
Fig 2. Body H, M, and L areas. Fig 1. Head and neck H and M areas.
(including perineal and perianal), hands, feet, nail units, ankles, and nipples/areola. Area M: cheeks, forehead, scalp, neck, jawline, and pretibial surface. Area L: trunk and extremities (excluding pretibial surface, hands, feet, nail units, and ankles).5,6 Skin cancers were deemed recurrent if patient notes documented a prior lesion treated in the same location or if the lesion was arising out of a previous treatment scar. Otherwise, skin cancers were considered primary. Patient characteristics (immunosuppression, transplantation, radiation, syndromes) The appropriate-use criteria note several patient characteristics including immunosuppression and transplantation status, previous irradiation to the skin cancer area, and history of genetic syndromes. These characteristics were documented using patient records. A patient was listed as immunosuppressed if they had a history of HIV, organ transplantation, hematologic malignancy, or pharmacologic immunosuppression. Treatment For skin biopsies occurring in the study period, patient records were followed up to the point of
treatment even if it was outside the study period. If a patient received multiple treatments, the initial treatment was counted. Treatment types included topical therapy (imiquimod, 5-fluorouracil), cryotherapy, curettage and electrodessication, excision, MMS, and radiation. MMS cases were referred to a University of Virginia Mohs surgeon. MIS/lentigo malignas were referred to a surgical oncologist, plastic surgeon, or otorhinolaryngologist for excision. Appropriate-use criteria After all patient and tumor characteristics were collected, each skin cancer was classified as appropriate, inappropriate, or uncertain based on the appropriate-use criteria. If the biopsy specimen was too superficial to determine histologic information, a classification was attempted using remaining information. For example, all BCCs in H areas are appropriate for MMS, regardless of subtype, based on the appropriate-use criteria. If the skin cancer was too superficial and could not be classified using the remaining information, it was excluded. Data analysis The percentages of skin cancers falling under appropriate, inappropriate, or uncertain designation were calculated. Percentages for patient
J AM ACAD DERMATOL
4 Blechman, Patterson, and Russell
Fig 3. Skin cancer selection. AUC, Appropriate-use criteria.
characteristics, tumor characteristics, and treatment types were determined. Standard statistics software showed that the sample size of this study can estimate the percentage of skin cancers appropriate for MMS for a similar population of any size with a 95% confidence level and a 63.0% confidence interval. A 2-tailed Fisher exact test was used to determine any statistically significant difference in MMS referral patterns between dermatologists and nondermatologist physicians. A P value equal to or less than .05 was deemed statistically significant. The statistical analysis was performed using software (SPSS Statistics 20.0, IBM Corp, Armonk, NY).
RESULTS From May 2011 to December 2011, a total of 1070 BCCs, SCCs, SCCs in situ (SCCISs), MIS/lentigo malignas, and other rare tumors were diagnosed by biopsy at the University of Virginia Health System. Of these, 11 biopsy specimens were too superficial to determine histologic characteristics necessary to make an appropriate-use designation (Fig 3). This left 1059 skin cancers with a breakdown of 667 (63.0%) BCCs, 244 (23.0%) SCCs, 108 (10.2%) SCCISs, 28 (2.6%) MIS/lentigo malignas, and 12 (1.1%) rare tumors. The rare tumors included 5 (0.5%) Merkel cell carcinomas, 2 (0.2%) angiosarcomas, 2 (0.2%) dermatofibrosarcoma protuberans, 1 (0.1%) leiomyosarcoma, 1 (0.1%) extramammary Paget disease, and 1 (0.1%) malignant fibrous histiocytoma. The location of these skin cancers varied among the H, M, and L areas with incidences of 33.4%, 25.9%, and 40.7%, respectively. Most pretreatment skin cancer sizes were in the range of 0.6 to 1.0 cm
(46.4%), followed by 1.1 to 2.0 cm (30.5%), less than or equal to 0.5 cm (16.0%), and greater than 2.0 cm (7.1%). The mean age was 68 years old with a range of 19 to 101 years. The majority (64.6%) of skin cancers were diagnosed in men. In 116 or 11.0% of all skin cancers, the patients were immunosuppressed. Of the 667 BCCs, 27.0% were aggressive subtypes, 21.4% nodular subtypes, and 30.1% superficial subtypes. Overall, 72.0% of skin cancers were appropriate for MMS based on the appropriate-use criteria. Likewise, 20.4% were deemed inappropriate and 7.6% uncertain. Among skin cancer types, SCCs had the highest proportion of skin cancers falling in the appropriate category at 90.6% (5.7% inappropriate, 3.7% uncertain). In the BCC group, 67.3%, 27.6%, and 5.1% were appropriate, inappropriate, and uncertain, respectively. In the SCCIS group, 63.0%, 16.7%, and 20.4% were appropriate, inappropriate, and uncertain, respectively. In the MIS group, 67.9%, 0.0%, and 32.1% were appropriate, inappropriate, and uncertain, respectively. Lastly, among all rare skin cancers, 50.0% were appropriate and 50.0% were uncertain. These percentages are summarized in Table I. The most common treatment types were excision (36.5%), MMS (29.7%), and electrodesiccation and curettage (22.2%). Other treatment statistics can be viewed in Table II. Among 1059 skin cancers, dermatologists diagnosed 987 and nondermatologist physicians identified 72. The proportion of skin cancers meeting appropriate use for MMS was 71.3% for dermatologists and 80.6% for nondermatologists, which was not a statistically significant difference (Table III).
J AM ACAD DERMATOL
Blechman, Patterson, and Russell 5
Table I. Appropriate-use criteria by skin cancer type Appropriate
BCC SCC SCCIS MIS/LM Other Total
449 221 68 19 6 763
(67.3) (90.6) (63.0) (67.9) (50.0) (72.0)
Inappropriate
184 14 18 0 0 216
(27.6) (5.7) (16.7) (0.0) (0.0) (20.4)
Uncertain
34 9 22 9 6 80
(5.1) (3.7) (20.4) (32.1) (50.0) (7.6)
Percentages of each skin cancer type are in parentheses. BCC, Basal cell carcinoma; LM, lentigo maligna; MIS, melanoma in situ; SCC, squamous cell carcinoma; SCCIS, squamous cell carcinoma in situ.
Table II. Treatment types Imiquimod, 5-fluorouracil Cryotherapy Electrodesiccation and curettage Excision Mohs micrographic surgery Radiation Other* Total
37 9 235 386 315 0 77 1059
(3.5) (0.8) (22.2) (36.5) (29.7) (0.0) (7.3) (100.0)
Percentages of all skin cancers are in parentheses. *Includes skin cancers that were completely removed by shave biopsy, that never received definitive treatment after shave biopsy, or for which the outcome was unknown.
More skin cancers appropriate for MMS were subsequently referred for MMS in the dermatologist group compared with the nondermatologist group (43.6% vs 13.8%, respectively). This difference was statistically significant (P = .000).
DISCUSSION The Medicare claims database from 2009 showed a 236% increase in MMS over 10 years. Part of this rise can be attributed to an increase in skin cancer prevalence. However, excisions and destructions increased by only 20% over the same time frame.8 Other reasons for the MMS surge include an increase in Mohs surgeons and a higher awareness by referring providers to the benefits of MMS. Although studies have shown MMS to have lower recurrence rates than other treatments,9 it is more expensive than some other treatment options, requires special training, requires special laboratory facilities, and is not available everywhere. Therefore, referring providers should understand which lesions are appropriate and inappropriate for MMS, and know treatment alternatives. In 1995, the American Academy of Dermatology proposed guidelines for MMS10; however, these lacked the level of detail
Table III. Mohs micrographic surgery referral patterns among dermatologists and nondermatologists Dermatologists
Skin cancers diagnosed Skin cancers meeting appropriate use Skin cancers referred for MMS Proportion of appropriate ultimately referred
Nondermatologists
P*
987 (100.0)
72 (100.0)
704 (71.3)
58 (80.6)
.103
307 (31.1)
8 (11.1)
.000
307/704 (43.6)
8/58 (13.8)
.000
Percentages of total skin cancers in each group are in parentheses. Dermatologists and nondermatologists had similar proportions of skin cancers meeting appropriate use for MMS, but more skin cancers appropriate for MMS were subsequently referred for MMS in the dermatologist group compared with the nondermatologist group. MMS, Mohs micrographic surgery. *Two-tailed Fisher exact test.
seen with the appropriate-use criteria published in 2012. In this study, tumor location was the most common factor used to establish which skin cancers met appropriate use. This finding was a result of 2 reasons. First, a large number of skin cancers occurred in H and M areas. Second, most skin cancers arising in H and M areas are designated appropriate for MMS by the appropriate-use criteria. Among the 72.0% of skin cancers that met appropriate use, 59.3% of them were in H and M areas. Under the appropriate-use criteria, all SCCs, BCCs, SCCISs, and MIS/lentigo malignas, except primary actinic keratosis (AKs) with focal SCCIS, are appropriate if located in the H area. H area skin cancers represented 33.4% of the sample size. Similarly, all skin cancers in the M area are appropriate except for superficial BCCs equal to or less than 5 mm in size and primary AKs with focal SCCIS, which accounted for another 25.8% of all lesions. Aggressive histologic features also contributed to a large proportion of appropriate lesions, especially for SCCs in which 90.6% were appropriate for MMS and 47.5% had aggressive subtypes. According to the appropriate-use criteria, any aggressive histologic feature for SCC makes it appropriate for MMS, regardless of all other characteristics. Among 244 SCCs, the most common aggressive features were Breslow depth 2 mm or more (30.7%), poor
6 Blechman, Patterson, and Russell
differentiation (8.2%), or both (2.0%). One limitation of the appropriate-use criteria is that it does not address all important tumor subtypes. For example, it does not include acantholytic or intravascular tumors as an aggressive subtype of SCC. Likewise, all BCCs are categorized into superficial, nodular, or aggressive subtypes even though there are many others. In these few situations, a University of Virginia dermatopathologist (J. W. P.) classified each skin cancer as best as possible using the information provided by the appropriate-use criteria. Skin cancers appropriate for MMS but not treated with MMS received some other treatment; electrodesiccation with curettage and excision were the most common other treatment types (Table II). The gap between 72% of skin cancers appropriate for MMS and only 29.7% referred for MMS was a result of various reasons such as patient comorbidities, scheduling constraints, travel constraints, and personal preference. Further, MMS was not used for MIS/ lentigo maligna although there were only 28 MIS/ lentigo malignas comprising less than 0.5% of our tumors. Also, skin cancers in this study were diagnosed and treated in 2011, which was before the appropriate-use criteria were published. Nevertheless, it implies that MMS might be underused. These statistics are not just specific to University of Virginia but studies from other institutions have cited similar MMS use rates of 30% to 35%.11-13 Dermatologists had a significantly higher MMS referral rate for appropriate skin cancers than that of nondermatologist physicians. Interestingly, of the 50 skin cancers diagnosed by nondermatologists that met appropriate use but were not referred for MMS, 35 or 70% were appropriate because of their locations in H and M areas. Likewise, in the dermatologist group, 73% of skin cancers meeting appropriate use but ultimately referred for other treatment methods were appropriate because of their location in H and M areas. Therefore, better understanding of H and M areas by both dermatologists and especially nondermatologists may lead to more skin cancers appropriate for MMS to be ultimately referred and treated with MMS. The purpose of this study was to apply the appropriate-use criteria to skin cancers at the University of Virginia Health System to determine the percent meeting appropriate use. These results provide insight into the potential demand for MMS at this institution but also other academic institutions. The percentages can also be applied on a national level to extrapolate the potential demand for MMS in the United States. By applying the 72.0%
J AM ACAD DERMATOL
appropriate-use figure to 2006 estimates of 3.5 million nonmelanoma skin cancers1 in the United States, 2.5 million would have been appropriate for MMS. This study’s sample size is large enough to make estimates of this magnitude within 3.0%. Currently, although most skin cancers may be appropriate for MMS, most were not treated by this modality. The ability to evaluate national trends may be limited because of differences in patient and tumor characteristics between academic and private practice settings. To address this issue, similar studies are warranted that encompass both academic and private practice patients. Also, because this study was retrospective, patient characteristics such as history of treatment and recurrence, radiation treatment to the area, genetic syndromes, and immunosuppression might have not been accounted for in some patients; this would underestimate the percent of cancers appropriate for MMS. Conclusion By applying the recently published appropriateuse criteria to histologically diagnosed skin cancers, it was this institution’s experience that 72.0% were appropriate for MMS. This percent was based not only on histology but also on anatomic location, pretreatment size, prior treatment, prior radiotherapy, and presence of immunologic problems. Tumor location was the most important factor, as almost all skin cancers in H and M areas were appropriate. Dermatologists were more likely than nondermatologist physicians to refer skin cancers that met appropriate use. Widespread understanding of H and M areas by both dermatologists and especially nondermatologists may lead to more skin cancers appropriate for MMS to be ultimately referred and treated with MMS. REFERENCES 1. Rogers HW, Weinstock MA, Harris AR, Hinckley MR, Feldman SR, Fleischer AB, Coldiron BM. Incidence estimate of nonmelanoma skin cancer in the United States, 2006. Arch Dermatol 2010;146:283-7. 2. Stern RS. Prevalence of a history of skin cancer in 2007: results of an incidence-based model. Arch Dermatol 2010;146: 279-82. 3. American Cancer Society. Cancer facts and figures 2012. Available from: URL:http://www.cancer.org/acs/groups/content/ @epidemiologysurveilance/documents/document/acspc-031941. pdf. Accessed December 10, 2012. 4. American Cancer Society. Cancer facts and figures 2002. Available from: URL:http://www.uhmsi.com/docs/CancerFacts& Figures2002.pdf. Accessed December 10, 2012. 5. Ad Hoc Task Force, Connolly SM, Baker DR, Coldiron BM, Fazio MJ, Storrs PA, Vidimos AT, et al. AAD/ACMS/ASDSA/ASMS 2012 appropriate use criteria for Mohs micrographic surgery: a report of the American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic
J AM ACAD DERMATOL
Surgery Association, and the American Society for Mohs Surgery. J Am Acad Dermatol 2012;67:531-50. 6. American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, Connolly SM, Baker DR, Coldiron BM, Fazio MJ, Storrs PA, Vidimos AT, et al. AAD/ACMS/ASDSA/ASMS 2012 appropriate use criteria for Mohs micrographic surgery: a report of the American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, and the American Society for Mohs Surgery. Dermatol Surg 2012;38:1582-603. 7. Anderson HL, Joseph AK. A pilot feasibility study of a rare skin tumor database. Dermatol Surg 2007;33:693-6. 8. Donaldson MR, Coldiron BM. Mohs micrographic surgery utilization in the Medicare population, 2009. Dermatol Surg 2012;38:1427-34.
Blechman, Patterson, and Russell 7
9. Thissen MR, Neumann MH, Schouten LJ. A systematic review of treatment modalities for primary basal cell carcinomas. Arch Dermatol 1999;135:1177-83. 10. Drake LA, Dinehart SM, Goltz RW, Graham GF, Hordinsky MK, Lewis CW, et al. Guidelines of care for Mohs micrographic surgery: American Academy of Dermatology. J Am Acad Dermatol 1995;33:271-8. 11. Donaldson MR, Coldiron BM. Mohs micrographic surgery utilization in the Medicare population, 2009. Dermatol Surg 2012;38:1427-34. 12. Gaston DA, Naugle C, Clark DP. Mohs micrographic surgery referral patterns: the University of Missouri experience. Dermatol Surg 1999;25:862-7. 13. Welch ML, Anderson LL, Grabski WJ. How many nonmelanoma skin cancers require Mohs micrographic surgery? Dermatol Surg 1996;22:711-3.