Cutaneous squamous cell carcinoma staging may influence management in users: A survey study

Abstract Purpose This study aims to determine whether there is consensus regarding staging and management of cutaneous squamous cell carcinoma (CSCC) across the various specialties that manage this disease. Materials and Methods A survey regarding CSCC high‐risk features, staging, and management was created and emailed to cutaneous oncology experts including dermatology, head and neck surgery/surgical oncology, radiation oncology, and medical oncology. Results One hundred fifty‐six (46%) of 357 invited physicians completed the survey. Depth of invasion (92%), perineural invasion (99%), histologic differentiation (85%), and patient immunosuppression (90%) achieved consensus (>80%) as high‐risk features of CSCC. Dermatologists were more likely to also choose clinical tumor diameter (79% vs. 54%) and histology (99% vs. 66%) as a high‐risk feature. Dermatologists were also more likely to utilize the Brigham and Women's Hospital (BWH) staging system alone or in conjunction with American Joint Committee on Cancer (AJCC) (71%), whereas other cancer specialists (OCS) tend to use only AJCC (71%). Respondents considered AJCC T3 and higher (90%) and BWH T2b and higher (100%) to be high risk and when they consider radiologic imaging, sentinel lymph node biopsy, post‐operative radiation therapy, and increased follow‐up. Notably, a large number of respondents do not use staging systems or tumor stage to determine treatment options beyond surgery in high‐risk CSCC. Conclusion This survey study highlights areas of consensus and differences regarding the definition of high‐risk features of CSCC, staging approaches, and management patterns between dermatologists and OCS. High‐risk CSCC is defined as, but not limited to, BWH T2b and higher and AJCC T3 and higher, and these thresholds can be used to identify cases for which treatment beyond surgery may be considered. Dermatologists are more likely to utilize BWH staging, likely because BWH validation studies showing advantages over AJCC were published in dermatology journals and discussed at dermatology meetings. Additional data are necessary to develop a comprehensive risk‐based management approach for CSCC.


| INTRODUCTION
Cutaneous squamous cell carcinoma (CSCC) is a rising epidemic, and while most tumors having an excellent prognosis, 4%-6% of tumors metastasize and 1.5%-2% lead to death. [1][2][3][4] Over the past several years, multiple CSCC staging systems and guidelines have been developed that vary in tumor assessment, including the AJCC 8 th Edition (AJCC), Brigham and Women's Hospital tumor staging system (BWH), University Hospital Tubingen (UHT) system, and National Comprehensive Cancer Network (NCCN) guidelines. 5 With no universal accepted definition of high-risk CSCC (HRCSCC), clear management guidelines are lacking. Point estimates of risk for recurrence and metastasis by tumor stage have varied between studies. [6][7][8] Subsequently, studies have found notable differences regarding the perioperative management of HRCSCC among Mohs surgeons, radiation oncologists, and head and neck surgeons. 9,10 Due to a paucity of data regarding the utility of nodal staging and adjuvant therapy, HRCSCC treatment is inconsistent and based on anecdotal experience.
This study aims to assess how physicians across the fields of dermatology, head and neck surgery/surgical oncology, radiation oncology, and medical oncology (latter three groups categorized as Other Cancer Specialists [OCS]) define HRCSCC and approach high risk, advanced, and/or metastatic CSCC. We hypothesized that there is variability between dermatologists and OCS with regard to HRCSCC definition, staging utilization, and management. Areas of consensus and areas needing clarification with future studies are identified.

| METHODS
A 25-question "Cutaneous SCC Staging Survey" ( Figure  S1) was developed by four of the authors (VAP, AJP, CDS, and STA) encompassing the management of HRCSCC. A list of dermatologists from the International Immunosuppression and Transplant Skin Cancer Collaborative and the American College of Mohs Surgery MohsAIQ registry, considered experts in cutaneous oncology, was created. 11 These individuals were contacted by email for OCS colleague referrals who frequently manage HRCSCC patients. Using these recommendations, a group of dermatologists and OCS were invited to participate.
Invited participants were emailed a personalized REDCap weblink to the survey and sent two reminders over 4 weeks. Incomplete surveys were excluded. The study was approved by the University of Texas at Austin Dell Medical School institutional review board.

| Statistical analysis
Descriptive summary statistics for survey responses were reported as frequency (percentage). Consensus was determined a priori at 80% or more of respondents in agreement. Comparisons between dermatologists and OCS were analyzed using Chi-square test (or Fisher's exact test for a cell-count ≤5). SAS version 9.4 was used for statistical analysis (SAS Institute Inc.). A two-sided p-value ≤0.05 was considered statistically significant.

| RESULTS
A total of 337 physicians (n = 163, 48% dermatologists and n = 174, 52% OCS) were invited to participate in the study. One hundred and fifty-six (46%) physicians completed the survey, of which 89 (57%) were dermatologists and 67 were (43%) OCS. Respondent characteristics, including specialty, number of years practicing, practice setting, and self-reported number of HRCSCCs managed in the previous 12 months are listed in Table 1. in dermatology journals and discussed at dermatology meetings. Additional data are necessary to develop a comprehensive risk-based management approach for CSCC.

K E Y W O R D S
American Joint Committee on Cancer Staging System, Brigham and Women Staging System, depth of tumor, high-risk cutaneous squamous cell carcinoma, high-risk tumor features, histologic differentiation, immunosuppression, perineural invasion, skin cancer, staging criteria, tumor diameter, tumor location

| High-risk features
Table S1 (ST1) lists the percentage of respondents selecting a feature as high risk.

| DISCUSSION
Accurate identification of HRCSCC helps to ensure we develop robust staging systems, identify those at high risk of poor outcomes, and provide precise treatments based on risk. The purpose of this study was to provide a multidisciplinary perspective regarding HRCSCC risk assessment and management. PNI, DI, HD, and immunosuppression reached consensus as high-risk factors by all experts. PNI ≥0.1 mm diameter, CTD (including >2 cm tumor diameter), and HD (including moderate differentiation, poor differentiation, spindle cell, desmoplasia, and infiltration) were more likely to be chosen by dermatologists than OCS. Our study demonstrates consensus in line with recent reports validating high-risk features and poor outcomes but also highlights the variability in definitions between experts. 9,10 The impact of such variability can be seen in the two current leading clinical trials evaluating adjuvant immunotherapy in HRCSCC; the inclusion criteria are different for both trials and do not follow either AJCC or BWH staging. 12,13 Results of these trials will be variable, making it difficult for clinicians or the NCCN to draw generalizable conclusions regarding adjuvant therapy. Nearly all respondents selected PNI and 81% specified that nerve diameter >0.1 mm was a significant cutoff. However, OCS were more likely to select any PNI as a risk factor, while dermatologists were more likely to specify PNI ≥0.1 mm diameter as high risk. This difference is critical as there is notable evidence in the dermatology literature that PNI of nerves <0.1 mm in diameter in the absence of other risk factors does not impact prognosis. 14,15 The lack of widespread discussion in the oncology literature may explain this variability. Dermatologists and Mohs surgeons also dedicate a large portion of their practice to HRCSCC, while most OCS are unlikely to have HRCSCC focus alone. While current staging and practice guidelines include PNI as a risk factor; ambiguity remains regarding management of the range of PNI and the variability of understanding true risk may lead to overtreatment in patients with smallcaliber PNI. 10,16 Immunosuppression is another feature with variable response. While clinical factors are typically not included in staging systems, there was a clear consensus of immunosuppression, and specifically SOTRs, as an important risk factor. This is unsurprising given that it is well known that SOTR who develop CSCC have worse outcomes than immunocompetent patients. 17,18 However, the identified risk from other immunosuppressive conditions was more variable between experts with OCS more likely to consider stem cell transplant and chronic systemic diseases as high risk compared to dermatologists. This is notable considering that evidence  of increased risk is minimal in these conditions. 19 The presence of immunosuppression needs clarity in regard to how to best incorporate this information with staging systems to make management decisions.
Uncertainty remains regarding what staging systems are best for risk assessment and management decisions for CSCC. Dermatologists utilize BWH staging more frequently than OCS, either on its own or in conjunction with AJCC, while OCS generally preferred AJCC. This is likely due to BWH being developed by dermatologists and is well discussed at dermatology conferences and in the dermatology literature. 6,7 OCS are likely accustomed to AJCC staging for other cancers and may thus prefer it. BWH has been shown to be superior to AJCC and many of the BWH risk factors were incorporated into the update from the seventh to eighth edition of AJCC, but with only 40% of respondents utilizing AJCC, it clearly needs further refinement before it is utilized widely. 11,20 Furthermore, 15% of respondents noted utilizing neither system highlighting a clear practice gap in the staging of CSCC.
There was consensus among those who used AJCC and BWH that tumor stage ≥AJCC T3 or ≥BWH T2b should be considered high-risk and radiologic imaging, SLNB, adjuvant radiation therapy, and increased follow-up would be considered in these tumors. This threshold to consider adjuvant therapy is not surprising given the number of studies that have shown high poor outcomes rates in these tumors. 21,22 However, the exact clinical scenarios and management options that should be employed in these patients are uncertain. Specifically, not all BWH T2b and AJCCT3 cases portend sufficient risk to warrant adjuvant therapy blindly, and data and guidelines vary regarding recurrence risk and management of this CSCC subset. Furthermore, there are a number of AJCC T2 and BWH T2a tumors that metastasize and lead to death. 19 This may explain why 43% of AJCC users denote T2 to be high risk and that AJCC T2/T3 tumors have a difficult overlap to discern between. The NCCN also recently created a new "very high risk" designation to unify management of a subset of tumors that have poor outcomes. (Table 3).
More concerning is that a large number of respondents are not using either system to drive their decision to perform radiologic imaging, SLNB, adjuvant radiation therapy, adjuvant systemic therapy, or increased follow-up. There is no clear consensus about how to utilize T staging to drive management despite evidence of the benefit of additional interventions in HRCSCC. 23,24 Uncertainty is expected given the inconsistent risk estimates by T stage. 19,20 Furthermore, the scarcity of staging system implementation in clinical practice may be compounded by the high incidence of CSCC, together with the T stage heterogeneity, as well as the potential lack of concern or knowledge about this tumor compared with melanoma. The underutilization of consistent and concrete guidelines signifies a major shortcoming in the care of patients with CSCC, especially HRCSCC, and needs urgent rectifying with accurate T-stage risk estimates and evidencebased treatment modalities by disease stage.
A strength of this study is that we surveyed experts in cutaneous oncology who treat a high number of HRCSCC, which allowed answers to be based on collective experiences. A limitation is that the panelists were identified from mainly academic tertiary referral centers and thus results may not be generalizable to all physician practice patterns. The survey was developed based on accepted literature evidence and current staging, but given the lack of complete understanding of CSCC, some answer choices in the survey may have been omitted. In these cases, many experts entered text that was not categorizable and thus not analyzed.
The results of this survey illustrate areas of consensus and equipoise between multispecialty experts that manage high risk and advanced CSCC. While there are many areas of agreement, there is significant room for communication and education. But most importantly, these results highlight the need for multidisciplinary consensus building and clinical trials to establish evidence-based criteria for risk stratification and management of HRCSCC.

CONFLICTS OF INTEREST
None.

AUTHORS CONTRIBUTIONS
Patel VA and Jambusaria-Pahlajani A involved in study conception and design. Patel VA and McCullum C carried out data collection. Patel VA, McCullum C, Sparks AD, Schmults CD, Arron ST, and Jambusaria-Pahlajani A involved in analysis and interpretation of results. Patel VA, McCullum C, Schmults CD, Arron ST, and Jambusaria-Pahlajani A carried out draft manuscript preparation. All authors reviewed the results and approved the final version of the manuscript.

ETHICAL STATEMENT
The research was conducted in accordance with the principles embodied in the Declaration of Helsinki and in accordance with local statutory requirements. All participants gave written informed consent to participate in the study.