Same Day Biopsy and Treatment of Non-Melanoma Skin Cancer in Patients with Field Cancerization: A Retrospective Chart Study

Background Patients with field cancerization will develop numerous superficial non-melanoma skin cancers (NMSCs). Treating patients with field cancerization can be challenging and burdensome due to the numerous non-melanoma skin cancers (NMSCs) they develop and the frequent dermatology visits required for biopsy and treatment. Objective The success rate of diagnosing and treating lesions suspicious for NMSCs on the same day is measured, immediately after biopsy. Methods We retrospectively reviewed records of patients with same day lesion diagnosis and curettage treatment to determine diagnostic accuracy, treatment failure, and number needed to treat to reduce a follow-up treatment. Results A total of 237 lesions underwent same day biopsy and treatment, of which the majority were NMSC (66%) or actinic keratosis (23%). Patients had at least 3 months and a median of 17 months follow-up. A total of 20 lesions either recurred or were deemed to require additional treatment. The number needed to treat (NNT) to prevent one follow-up treatment was 1.3. Limitations: sample size limited ability to determine risk factors for treatment failure. Conclusion Simultaneous diagnosis and treatment of superficial NMSCs is a successful way of improving efficiency and patient satisfaction.


Introduction
Sun damage from excessive sun-exposure causes changes to the skin which leads to formation of keratinocyte neoplasms (KNs) such as precancerous actinic keratoses (AKs) and subsequent non-melanoma skin cancer (NMSC) such as squamous cell carcinoma [1]. While AKs are not thought to be precursor lesions to basal cell carcinoma (BCC), they are a marker of signifcant sun damage which is also thought to cause BCC. Over time, sun-damaged keratinocytes develop DNA mutations. Tese cells then give rise to similarly affected "progeny" cells which leads to "actinic feld cancerization" in areas of chronic, unprotected sun-exposure such as scalp, face, and sun-exposed trunk and extremities. Once a patient develops feld cancerization on a particular anatomic site, they are at signifcantly increased risk of developing numerous superfcial NMSCs within the afected area(s) [2,3]. Normally, patients come to see dermatologists where concerning areas are biopsied and sent to a dermatopathologist for analysis. If results show NMSC, patients will return for treatment of one or several lesions by curettage or surgical excision [4]. Certain lesions may also be amenable to topical treatment. It is common to biopsy multiple sites, have the patient return for several visits to address these NMSCs, and then have the patient return again within a few months or even weeks to repeat the process [3]. Te cycle gives rise to "treatment fatigue" for patients who may actually stop coming for routine screening visits. It also leads to reduced access to other patients who could use these clinic visits. Finally, it results in a signifcant cost to our health care system. Te average number of individuals treated for skin cancer in the United States rose from 3.4 to 4.9 million between 2002-2006 and 2007-2011 while the cost for treating skin cancer rose 126%, from 3.6 to 8.1 billion dollars [5]. Annual spending for skin cancers increased more than for any other cancer [6].
Many dermatologists have recognized these challenges and have anecdotally biopsied and treated suspected NMSCs at the time of the visit to help address these concerns. Tis same day biopsy and treatment serve to accomplish several things such as (i) increase efciency: by addressing concerning lesions at the time of presentation, it would eliminate the need for subsequent visit, (ii) improve patient satisfaction and reduce "treatment fatigue, and (iii) reduce cost to our health care system: if treatment is performed at time of the biopsy, providers cannot bill the cost of both the biopsy and the treatment. If the lesion is a skin cancer, the provider can only bill the cost of the treatment and not the biopsy. Conversely, if the lesion is benign, only the biopsy can be billed to insurance. One barrier to utilizing this technique is concern about efcacy of treatment as well as overtreatment of benign lesions. We sought to examine these concerns in addition to other areas of interest including efcacy related to skin cancer type, location, and immune status.

Patient Selection and Treatment.
Tis study was approved by the UNC institutional review board. Patients from UNC Dermatology Clinic were included in this chart review if they were found to have lesions concerning for superfcial skin cancers such as sBCC, SCCis, and early nodular BCC and were amenable to this treatment approach. Lesions were anesthetized using 1% lidocaine with epinephrine (1 : 100,000), and shave biopsy was performed. Immediately following biopsy, sites were treated with curettage with or without light electrodessication. Patients were subsequently monitored with regular follow-up.

Statistical Analysis.
Medical records were reviewed for demographic information, immunosuppression status, history of NMSC, and lesion characteristics, including diagnosis, location, and size. Data were stored in an online secure database, RedCap [7] and analyzed with Stata, version 16 (StataCorp LLC, College Station, TX). Primary outcomes were diagnostic accuracy and treatment failure. Diagnosis was considered accurate if histopathology identifed a keratinocytic neoplasm (NMSC or AK). All other lesions were classifed as benign. Treatment failure was defned as tumors that either recurred or were deemed to require additional treatment after initial histopathological analysis was performed. Secondary outcomes were predictive relationships between patient and lesion characteristics with treatment failure as assessed by univariate and multivariate logistic regression, correlation between patient and lesion characteristics with treatment failure as assessed by X 2 tests, unpaired t-test, or Fisher exact tests and lastly calculating the number needed to treat (NNT) to prevent follow-up treatment.

Results
From July 2016 to October 2019, 63 patients received same day biopsy and treatment with curettage of 244 lesions and were followed for at least 3 months. Lesion characteristics are summarized in Table 1. Majority of lesions occurred on male patients (81%) at a median age of 73 years. In total, 211 lesions (89%) were pathologically confrmed as KN (Table 2) demonstrating either squamous cell carcinoma in situ (SCCis; n � 71), invasive squamous cell carcinoma (iSCC; n � 27), basal cell carcinoma (BCC; n � 62), or actinic keratosis (AK or hypertrophic AK (HAK); n � 57). Te 28 benign diagnoses are listed in Table 2.
Median follow-up time was 32 months (interquartile range 20-40.5), during which time 23 lesions (10.6% of 216 KNs) failed treatment, either by recurring (n � 6) or requiring additional treatment (n � 17) due to incomplete resolution of the initial lesion. Failure occurred in 10 of 28 (36%) iSCC, 8 of 68 (12%) SCCis, and 5 of 64 (8%) BCC. No treatment failures occurred in any of the 56 AK ( Table 3). Seven of the ten iSCC failures were in lesions ranging from 10 to 19 mm versus iSCC less than 10 mm (2 of 10). One iSCC was missing size information. Tere were no iSCC lesions greater than 19 mm in this study. Of the 12 SCCis and BCC lesions treated that were greater than 19 mm, only four SCCis lesions were considered treatment failures.
Univariate analysis of a priori patient and lesion characteristics from Table 1

Journal of Skin Cancer
Using the data from all lesions in this sample as an estimate for the rates of benign, cancerous, and precancerous lesions in biopsied specimens, we estimate that 89% of biopsies would require follow-up treatment after being identifed as either NMSC or AK. Same day treatment reduces the rate of follow-up treatment to 10.6%. Tis is an absolute risk reduction of 78% and a NNT of 1.3. Excluding actinic keratosis in the calculations gives a more conservative NNT of 2.0 given a baseline follow-up treatment rate of 66% reduced to 14%.

Conclusion
Efciency, cost savings, and patient satisfaction are becoming increasingly valued metrics in medicine. In eforts to save patients time, decrease cost, and increase efciency, we have been treating lesions suspicious for superfcial NMSCs with curettage immediately after biopsy. Risk of "overtreatment" of benign lesions and "under treatment" of invasive lesions was reviewed prior to biopsy. When presented with the option of curettage at time of biopsy, we observed that patients preferred this to waiting for results and scheduling a return visit for treatment (although this was not conducted as a formal survey). An important prerequisite for performing this technique is having a high pretest probability in identifying superfcial KN. Terefore, we examined success and failure rates as the ability to distinguish superfcial NMSC from benign lesions. Tere is much variability in the literature describing rates of NMSC diagnosis in patients undergoing biopsy concerning for possible skin cancers. Some reports describe approximately 50% rate of diagnosing NMSC [8] to over 85% [9]. We report  a 66% rate of NMSC diagnosis (BCC or SCC) increasing to 89% when including AK and HAK. Another component of measuring success had to do with how well we were able to eliminate the need for further treatment, whether it is cryotherapy, topical therapy, curettage, or excision. By performing curettage immediately after biopsy, the intention was to eliminate the need for additional treatment or visits to confrm clearance if a topical agent was utilized. We found a large reduction in need for further treatment with a NNT to reduce one follow-up treatment of 1.3 to 1.9. In other words, same day biopsy and curettage of 100 lesions would over treat 11 benign lesions while sparing 51 to 78 lesions from needed follow-up treatment.
Curettage with or without electrodessication has been employed by dermatologists since the beginning of the 20 th century [10]. Tere are numerous reports describing success and recurrence rates in treating NMSC. An important study by Silverman et al. from 1991 measured recurrence of BCC treated by ED&C from 1955 to 1982 at NYU [11]. Tis study shaped many of the modern principles and guidelines regarding the use of curettage with and without electrodessication in dermatology. Higher rates of recurrence were found for BCCs treated on the nose, chin, ear, and periorifcial locations while lowest risks of recurrence were found on the trunk, extremities, and neck. Lesions less than 1 cm showed a 9-10% 5-year recurrence (all sites). However, recurrence rates in lower risk areas, regardless of lesion diameter, were only between 2 and 4%. Another important consideration is subtype of BCC. Infltrative and morpheaform subtypes are not amenable to ED&C, and it is estimated that ∼13% of BCCs show either of these subtypes [12]. Same day biopsy and curettage for BCC in our analysis showed an 8% failure rate as measured by recurrence or need for additional treatment which is consistent with prior reports. Tere were no infltrative or morpheaform BCCs biopsied or treated in this analysis.
Many studies have demonstrated curettage with or without electrodessication to be a valuable treatment option for SCCis with estimates of success averaging between 93 and 98% [13]. Most reports exclude SCCis with follicular or adnexal involvement since cure rates are much lower for this subtype. Te data presented in our analysis showed an 88% overall success rate in treating SCCis. An additional concern was presence of iSCC found in previously biopsied SCCis discovered during standard excision or Mohs which has been described [14]. While this has been described, there were no such cases discovered in this analysis.
Invasive SCC is typically treated surgically by standard excision or Mohs micrographic surgery. Nonetheless, numerous studies including recent reports have demonstrated excellent success treating lesions with curettage. A recent report showed a 97% success rate using curettage alone to treat iSCC [15]. Success rate in our analysis was 64%. It is important to note that 4 lesions that were not adequately treated by same day biopsy and curettage were treated by Mohs.
While about a third of lesions treated were on immunocompromised patients, almost half of treatment failures came from lesions on these patients. However, in statistical analyses, immunocompromised status did not confer signifcant increased risk of treatment failure. All of these lesions that failed treatment were either invasive SCC or SCCis. It is well-known that immunosuppressed patients are at increased risk of recurrence as well as distant spread so the decision was made in many of these cases to refer to Mohs [16].
Treatment success was signifcantly diferent between lesion locations, demonstrating the highest success rate on the trunk (98%) compared to the face, ear, or neck locations with a success rate of 77%. Unsurprisingly, success rate varied signifcantly by histology. Same day curettage showed a 92% success rate for BCC (combining nodular and superfcial types) while success rates for SCCis and iSCC were 88% and 64%, respectively. Importantly, there was no signifcant diference between success rates in welldiferentiated vs. moderately diferentiated SCCs treated at the time of biopsy (55% vs. 56%, data not shown). Additionally, there was no signifcant diference between success rates in superfcial or nodular BCC (95% vs. 90%, data not shown). While success rates did not signifcantly vary by lesion size, there was a trend towards lower success with larger lesions. However, it is worth noting that success rate for SCCis less than 2 cm was 93% (56/60). Success rate for SCCis lesions measuring greater than 2 cm was only 50%. Success rate for BCCs less than 2 cm was 91% (53/58).
Tis analysis would have beneftted from more patients and a greater number of tumors. Few treatment failures limit the interpretation of statistical analyses to evaluate potential predictive variables or confounders for treatment failure.
Overall, there was a high success rate in treating lesions suspicious for NMSC immediately following biopsy; the vast majority (91%) of the tumors showed no evidence of recurrence with a minimum of 3 months follow-up. Tis practice shows a large reduction in need for follow-up appointments at the cost of a few benign lesions undergoing treatment. Te decision to same day treatment should be a mutual decision between provider and patient. Tis analysis adds data to inform that discussion.
We conclude that this approach should be considered for treating clinically diagnosed NMSC lesions on relatively small-and superfcial-appearing lesions on immunocompetent patients. Caution should be exercised in treating lesions on the face, larger lesions, or tumors in immunocompromised patients with curettage. Also, this method is not appropriate for lesions that clinically are suspected to have an infltrative or deeply invasive growth pattern. Providers who want to carry out this technique should always ensure regular follow-up and monitor the lesions routinely. Altogether, our hope is that this study demonstrates the feasibility of treating superfcial skin cancers at the time of biopsy to improve patient satisfaction, efciency, and cost savings to our healthcare system.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request. 4 Journal of Skin Cancer