This cohort study assesses the association between the presence of immunosuppression and disease-specific outcome in patients with cutaneous squamous cell carcinoma.
Key Points
Question
What is the association between immune status and disease-specific outcome in patients with cutaneous squamous cell carcinoma?
Findings
In this cohort study of 796 patients with cutaneous squamous cell carcinoma, patients with immunosuppression had twice the risk for disease-specific death compared with those without immunosuppression, even after adjustment for age, history of skin cancer, recurrence or persistence of disease, stage, and treatment.
Meaning
These findings suggest that immunosuppression is an independent risk factor associated with a worse outcome in patients with cutaneous squamous cell carcinoma.
Abstract
Importance
Patients with immunosuppression have a higher incidence of cutaneous squamous cell carcinoma (cSCC) and often present with more aggressive, multifocal disease.
Objectives
To determine the risks for mortality in patients with cSCC and immunosuppression compared with nonimmunosuppression and to compare the difference in mortality risk based on the cause of immunocompromise.
Design, Setting, and Participants
This retrospective cohort study of patients with cSCC of the head and neck recruited participants from a tertiary cancer care center. Patients who underwent no treatment, wide local excision, or biopsy of the lesions were eligible for inclusion from January 1, 1995, to September 30, 2015. Data were analyzed from March 21, 2018, to April 4, 2019.
Exposures
Immunocompromise, defined as having solid organ transplant, stem cell transplant, hematopoetic malignant disease, autoimmune disease requiring treatment with immunosuppressive therapy, type 1 or 2 diabetes treated with insulin, HIV or AIDS, or other hematoproliferative disorder.
Main Outcomes and Measures
Patients were divided into 2 groups according to their immune status (immunosuppression vs no immunosuppression). The primary outcome measure was disease-specific survival. A Cox proportional hazards regression model was used to determine the association of immune status with disease outcome.
Results
A total of 796 patients (680 men [85.4%]; median age, 69 [range, 27-98] years), including 147 with and 649 without immunosuppression (IS and non-IS groups, respectively), constituted the final cohort. In the IS group, 77 (52.4%) had diabetes, 39 (26.5%) had lymphoma or leukemia, 25 (17.0%) had an organ or stem cell transplant, and 3 (2.0%) had HIV. Five-year disease-specific survival was 68.2% in the IS group compared with 84.1% in the non-IS group (difference, 15.9%; 95% CI, 3.5%-27.4%). Immunosuppression was independently associated with worse disease-specific survival (hazard ratio, 2.32; 95% CI, 1.53-3.50).
Conclusions and Relevance
This study’s findings suggest that immunosuppression is independently associated with a worse outcome in cSCC, with a 2.32 times increased risk of disease-specific death after adjusting for age, history of skin cancer, recurrent or persistent disease status, disease stage, and treatment.
Introduction
Cutaneous squamous cell carcinoma (cSCC) is the second most common malignant neoplasm of the skin, representing approximately 20% of all cases.1 The head and neck region is the most common site of cSCC, posing unique challenges due to the cosmetic and functional constraints of treatment.2 Immunosuppression is a major known risk factor for cSCC of the head and neck (cSCCHN). Patients with immunosuppression are at a 65- to 100-fold increased risk of developing cSCC.3 Although the incidence of basal cell carcinoma is 4:1 in the immunocompetent population, this ratio is reversed in the immunosuppressed population, with cSCC being 4 times more common than basal cell carcinoma.4
Patients with a history of immunosuppression not only develop cSCCHN more frequently, but the skin cancers they develop appear to behave more aggressively. Patients with immunosuppression have a higher propensity for poorly differentiated disease, lymphovascular invasion, and extracapsular extension compared with those without immunosuppression.5,6 Patients with immunosuppression are also more likely to present with early dermal invasion and multifocal disease, have thicker primary tumors, and develop nodal disease.4,7,8,9
Surgical excision is considered the primary treatment in cSCCHN. In patients with aggressive disease, such as those with a history of immunosuppression, adjuvant therapies can play a major role. Options for incorporating systemic therapy in the management of cSCCHN have expanded with the introduction of targeted therapies and immunotherapy, underscoring an increased need to better understand the effects of immunosuppression on the outcome of cSCCHN. The aim of this study is to evaluate survival after treatment of cSCCHN in patients with immunosuppression compared with patients without immunosuppression.
Methods
This retrospective cohort study included patients with cSCCHN treated at the University of Texas MD Anderson Cancer Center (MDACC) from January 1, 1995, to September 30, 2015. Approval was obtained from the MDACC institutional review board, and a waiver of informed consent was granted due to the retrospective nature of deidentified data collection. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Patients 18 years or older were eligible for the study if they had a pathologically confirmed diagnosis of cSCCHN. Patients were excluded if they received any treatment other than biopsy or local excision before presentation to MDACC. Other exclusion criteria consisted of premalignant lesions only and follow-up of less than 1 month. Demographic, clinical, treatment, and pathologic data were collected from the electronic medical records. Race/ethnicity was determined by the patient. All of the information obtained from medical record review was maintained in Research Electronic Data Capture (REDCap), a secure, web-based database application.
Patients were divided into 2 groups based on immune status. Patients were considered to have immunosuppression (IS group) if they had (1) solid organ transplant, (2) stem cell transplant, (3) hematopoetic malignant disease (eg, lymphoma or leukemia), (4) autoimmune disease requiring treatment with immunosuppressive therapy, (5) type 1 or 2 diabetes treated with insulin, (6) HIV or AIDS, or (7) other hematoproliferative disorder. All other patients were considered not to have immunosuppression (non-IS group). The primary outcome measure was disease-specific survival (DSS), which was defined as the time from initial presentation to MDACC to death due to cSCCHN or loss to follow-up. Secondary outcome measures were overall survival (OS), defined as the time from initial presentation to MDACC to death due to any cause or loss to follow-up, and recurrence-free survival, which was defined as the time from initial presentation to MDACC to recurrence of disease.
Data were analyzed from March 21, 2018, to April 4, 2019. Descriptive statistics, including median and range, were determined for all baseline characteristics. Effect size metrics were reported as absolute differences between means for continuous variables and differences in proportions for categorical variables, along with their 95% CIs. Survival probabilities were calculated using the Kaplan-Meier method. The univariate and multivariate Cox proportional hazards regression models for OS and DSS were used to explore possible covariates associated with survival. The forward selection method was used for covariate selection in the multivariate model to include all variables with statistically significant associations on univariate analysis. To understand the association of different causes of immunosuppression with DSS, subgroup analysis was completed using the same covariates identified by the overall model for adjustment. In each of these comparisons, individuals with a specific cause of immunosuppression were compared with patients without immunosuppression. All tests were 2-tailed. All statistical analyses were completed using STATA, version 14.2 (StataCorp LLC).
Results
Of the 796 patients included in the final cohort (116 women [14.6%] and 680 men [85.4%]; median age, 69 [range, 27-98] years), 147 (18.5%) were in the IS group, and 649 (81.5%) were in the non-IS group (Table 1). No difference was found in the median age, racial/ethnic distribution, and follow-up time between the IS and non-IS groups. The IS group included more men (136 [92.5%] compared with 544 [83.8%] in the non-IS group; difference, 8.7% [95% CI, 3.6%-13.8%]). In the IS group, more patients had a history of nonmelanoma skin cancer (109 [74.1%] vs 418 [64.4%] in the non-IS group; difference, 9.7% [95% CI, 1.7%-17.7%]).
Table 1. Baseline Characteristics of All Patients With Cutaneous Squamous Cell Carcinoma According to Immune Status.
| Characteristic | Study Groupa | Effect Size (95% CI), %b | |
|---|---|---|---|
| No Immunosuppression (n = 649) | Immunosuppression (n = 147) | ||
| Age, median (range), y | 70 (27-97) | 69 (31-98) | 12.2 (−2.18 to 2.18) |
| Sex | |||
| Male | 544 (83.8) | 136 (92.5) | 8.7 (3.6 to 13.8) |
| Female | 105 (16.2) | 11 (7.5) | NA |
| Race/ethnicity | |||
| White | 605 (93.2) | 140 (95.2) | 2.0 (−2.9 to 6.0) |
| Black | 5 (0.8) | 0 | −0.8 (−1.4 to −0.1) |
| Hispanic | 35 (5.4) | 7 (4.8) | −0.6 (−4.5 to 1.9) |
| Other | 4 (0.6) | 0 | 0.6 (0.01 to 1.2) |
| History of skin cancer | |||
| Yes | 418 (64.4) | 109 (74.1) | 9.7 (1.7 to 17.7) |
| No | 231 (35.6) | 38 (25.9) | NA |
| Recurrent or persistent disease | |||
| Yes | 309 (47.6) | 76 (51.7) | 4.1 (−4.9 to 13.0) |
| No | 340 (52.4) | 71 (48.3) | NA |
| T category | |||
| T0 | 73 (11.2) | 10 (6.8) | −4.4 (−9.2 to 0.3) |
| T1 | 266 (41.0) | 59 (40.1) | −0.9 (−9.6 to 7.9) |
| T2 | 180 (27.7) | 47 (32.0) | 4.2 (−4.1 to 12.5) |
| T3 | 59 (9.1) | 17 (11.6) | 2.5 (−3.1 to 8.1) |
| T4 | 71 (10.9) | 14 (9.5) | −1.4 (−6.7 to 3.9) |
| N category | |||
| N0 | 488 (75.2) | 117 (79.6) | 4.4 (−2.9 to 11.7) |
| N1 | 60 (9.2) | 13 (8.8) | −0.4 (−5.5 to 4.7) |
| N2 | 97 (14.9) | 17 (11.6) | −3.4 (−9.2 to 2.5) |
| N3 | 2 (0.3) | 0 | −0.3 (−0.7 to 0.1) |
| NX | 2 (0.3) | 0 | −0.3 (−0.7 to 0.1) |
| Overall AJCC-8 stage | |||
| I | 240 (37.0) | 58 (39.5) | 2.5 (−6.3 to 11.2) |
| II | 139 (21.4) | 36 (24.5) | 3.1 (−4.6 to 10.7) |
| III | 92 (14.2) | 23 (15.6) | 1.5 (−5.0 to 7.9) |
| IV | 156 (24.0) | 29 (19.7) | −4.3 (−11.5 to 2.9) |
| Unknown | 22 (3.4) | 1 (0.7) | −2.7 (−4.6 to −0.8) |
| ECOG performance statusc | |||
| 0 | 53 (8.2) | 5 (3.4) | −4.8 (−8.4 to −1.2) |
| 1 | 330 (50.8) | 45 (30.6) | −20.2 (−28.6 to −11.9) |
| 2 | 15 (2.3) | 2 (1.4) | −1.0 (−3.2 to 1.3) |
| 3 | 2 (0.3) | 5 (3.4) | 3.1 (0.1 to 6.1) |
| 4 | 1 (0.2) | 0 | −0.2 (−0.5 to 0.1) |
| Unknown | 248 (38.2) | 90 (61.2) | 23.0 (14.3 to 31.7) |
| Prior treatment | |||
| None | 268 (41.3) | 50 (34.0) | −7.3 (−15.8 to 1.3) |
| Surgery only | 381 (58.7) | 97 (66.0) | NA |
| Chemotherapy only | 0 | 0 | NA |
| Radiotherapy only | 0 | 0 | NA |
| Surgery and adjuvant therapy | 0 | 0 | NA |
| Radiotherapy and systemic therapy | 0 | 0 | NA |
| Follow-up, median (range), mo | 32.5 (1-258) | 32 (1-258) | 7.40 (−8.18 to 22.98) |
Abbreviations: AJCC-8, American Joint Committee on Cancer, Eighth Edition; ECOG, Eastern Cooperative Oncology Group; NA, not applicable.
Unless otherwise indicated, data are expressed as number (percentage) of patients. Percentages have been rounded and may not total 100.
Given as mean difference presented for continuous variable; absolute difference, for categorical variables.
Scores range from 0 to 4, with higher scores indicating poorer overall function.
The most common T category at presentation included T1 and T2 (IS group, 106 [72.1%]; non-IS group, 446 [68.7%]; difference, 3.4% [95% CI, −4.7% to 11.5%]), and most patients had no nodal disease at presentation (IS group, 117 [79.6%]; non-IS group, 488 [75.2%]; difference, 4.4% [95% CI, −2.9% to 11.7%]). Overall American Joint Committee on Cancer stage distribution was similar between the 2 groups at presentation. None of the patients had prior chemotherapy or radiotherapy for their cSCC, although 97 patients [66.0%] in the IS group and 381 [58.7%] in the non-IS group had prior surgery.
The most common form of immunosuppression in this cohort was diabetes (n = 77 [52.4%]), followed by lymphoma or leukemia (n = 39 [26.5%]), organ and stem cell transplant (n = 25 [17.0%]), autoimmune disease requiring treatment (n = 6 [4.1%]), and HIV (n = 3 [2.0%]). Five patients had other causes of immunosuppression, including myelofibrosis, multiple myeloma, myelodysplasia, chronic immune demyelinating polyneuropathy, and lymphocytic meningitis. Eight patients had more than 1 cause of immunosuppression.
Treatment of Patients With cSCC
Most patients underwent surgery at MDACC (IS group, 119 [81.0%]; non-IS group, 533 [82.1%]; difference, −1.2% [95% CI, −8.2% to 5.8%]), with more than half of patients being treated with postoperative radiotherapy (IS group, 85 [57.8%]; non-IS group, 355 [54.7%]; difference, 3.1% [95% CI, −5.7 to 12.0%]) and some with chemotherapy (IS group, 17 [11.6%]; non-IS group, 113 [17.4%]; difference, −5.8% [95% CI, −11.8% to 0.1%]). There was no difference in selection of treatment between the IS and non-IS groups.
Survival According to Immune Status
Patients in the IS group had significantly lower DSS compared with the non-IS group at 3 years (75.8% vs 87.1%; difference, 11.3% [95% CI, 2.0%-20.7%]), 5 years (68.2% vs 84.1%; difference, 15.9% [95% CI, 3.5%-27.4%]), and 10 years (62.4% vs 83.2%; difference, 20.8% [95% CI, 8.7%-34.0%]) (Figure 1A). Overall survival was also significantly reduced in the IS group at 3 years (56.2% vs 70.4%; difference, 14.2% [95% CI, 3.9%-24.9%]), 5 years (36.5% vs 58.9%; difference, 22.4% [95% CI, 11.2%-33.3%]), and 10 years (17.1% vs 40.4%; difference, 23.3% [95% CI, 12.2%-33.6%]) (Figure 1B). There was no meaningful difference between groups in recurrence-free survival at 3 years (IS group, 79.6%; non-IS group, 87.7%; difference, 8.1% [95% CI, −1.6% to 16.5%]) and 5 years (IS group, 75.8%; non-IS group, 84.9%; difference, 9.1% [95% CI, −0.8% to 19.0%]).
Figure 1. Kaplan-Meier Survival Curves by Immunosuppression Group.
Data are shown for disease-specific survival (A) and overall survival (B) among patients with and without chronic immunosuppression and cutaneous squamous cell carcinoma of the head and neck.
Patterns of Failure
There was no meaningful difference in either cohort between the types of recurrence. Patients were most likely to develop a second primary tumor (IS group, 23 [15.6%]; non-IS group, 111 [17.1%]; difference, −1.5% [95% CI, −8.0% to 5.1%]), followed by a local recurrence (IS group, 18 [12.2%]; non-IS group, 66 [10.2%]; difference, 2.1% [95% CI, −3.7% to 7.9%]). Regional recurrences occurred in 9 patients (6.1%) in the IS group and 40 (6.2%) in the non-IS group, whereas distant recurrences occurred in 6 (4.1%) in the IS group and 25 (3.9%) in the non-IS group, with no statistical variance between groups (differences, 0.0% [95% CI, −4.3% to 4.3%] and 0.2% [95% CI, −3.3% to 3.8%], respectively).
Univariate and Multivariate Analyses
Univariate analysis found that presence of IS conferred a 2.15 increased risk of death due to disease (95% CI, 1.44-3.21). Age (per year; hazard ratio [HR], 1.02; 95% CI, 1.01-1.04), history of skin cancer (HR, 2.28; 95% CI, 1.44-3.60), recurrent skin cancer (HR, 2.25; 95% CI, 1.53-3.31), advanced T3 to T4 category (HR, 2.68; 95% CI, 1.82-3.94), and advanced nodal category (N2-N3; HR, 3.33; 95% CI, 2.20-5.03) were also associated with worse DSS. Multivariable analysis was performed, including adjustment for age, history of skin cancer, recurrent or persistent disease, T and N categories, and immune status. Immunosuppression remained independently associated with DSS (HR, 2.32; 95% CI, 1.53-3.50). Other covariates included advanced age (HR, 1.02; 95% CI, 1.00-1.04) and advanced disease stage (HR for T3-T4 category, 2.45 [95% CI, 1.66-3.63]; HR for N2-N3 category, 2.45 [95% CI, 1.66-3.63]) (Table 2).
Table 2. Univariate and Multivariate Cox Proportional Hazards Regression Models for Disease-Specific Survival.
| Characteristic | HR (95% CI) | |
|---|---|---|
| Univariate | Multivariate | |
| Immunosuppression | ||
| Yes | 2.15 (1.44-3.21) | 1.95 (1.31-2.93) |
| No | ||
| Age, ya | 1.02 (1.01-1.04) | 1.02 (1.00-1.04) |
| Sex | ||
| Male | 1 [Reference] | 1 [Reference] |
| Female | 0.70 (0.39-1.28) | NA |
| History of skin cancer | ||
| No | 1 [Reference] | 1 [Reference] |
| Yes | 2.28 (1.44-3.60) | 1.52 (0.82-2.81) |
| Recurrent or persistent disease | ||
| No | 1 [Reference] | 1 [Reference] |
| Yes | 2.25 (1.53-3.31) | 1.35 (0.80-2.29) |
| T category | ||
| T0, T1, and T2 | 1 [Reference] | 1 [Reference] |
| T3 and T4 | 2.68 (1.82-3.94) | 2.45 (1.66-3.63) |
| N category | ||
| N0 | 1 [Reference] | 1 [Reference] |
| N1 | 1.41 (0.75-2.67) | 1.34 (0.70-2.54) |
| N2 and N3 | 3.33 (2.20-5.03) | 2.45 (1.66-3.63) |
Abbreviations: HR, hazard ratio; NA, not applicable.
Increase in hazard per year.
Subgroup Analyses by Type of Immunosuppression
Subgroup analysis, using the same final model outlined above and adjusting for age, history of skin cancer, recurrent or persistent disease, T category, and N category, is depicted in Figure 2. Patients with a diagnosis of HIV or AIDS had the highest risk of death due to disease (HR, 7.73; 95% CI, 1.01-59.16). This risk was followed by patients with hematopoietic malignant disease (HR, 3.50; 95% CI, 1.93-6.35) and organ and stem cell transplant (HR, 2.49; 95% CI, 1.05-5.86). Diagnosis with diabetes treated with insulin was not independently associated with worse DSS when considered alone as a cause of immunosuppression (HR, 1.28; 95% CI, 0.72-2.28). However, when Kaplan-Meier survival analysis was performed, we found no meaningful difference between cause of immunosuppression and DSS (HR for hematopoietic malignant disease, 1.88 [95% CI, 0.87-4.08]; HR for organ and stem cell transplant, 1.44 [95% CI, 0.55-3.81]) or OS (HR for hematopoietic malignant disease, 1.46 [95% CI, 0.90-2.36]; HR for organ and stem cell transplant, 1.06 [95% CI, 0.57-1.99]), with diabetes as the reference value (Figure 3).
Figure 2. Forest Plot for Association of Cause of Immunosuppression With Disease-Specific Survival.
Analysis used a subgroup multivariate Cox proportional hazards regression model and adjusted for age, history of skin cancer, recurrent or persistent disease, T and N categories, and treatment (surgery, radiotherapy, or chemotherapy). HR indicates hazard ratio.
Figure 3. Kaplan-Meier Survival Curves by Immunosuppression Group and Cause of Immunosuppression.
Disease-specific survival (A) and overall survival (B) among patients with and without chronic immunosuppression and cutaneous squamous cell carcinoma treated according to cause of immunosuppression. DM indicates diabetes mellitus.
Discussion
Cutaneous SCCHN in patients with immunosuppression is associated with poor outcomes and aggressive behavior when compared with the immunocompetent population.4,5,6,7,8,9 In the present study, we compared disease-specific outcomes in a cohort of patients with immunosuppression and a variety of causes vs a cohort without immunosuppression. The cohort of 147 patients in the IS group with cSCCHN constitutes a large, single-institutional study for determining treatment outcomes in patients with cSCCHN and immunosuppression vs no immunosuppression.
We found no meaningful differences in treatment between the IS and non-IS cohorts, with just more than half of patients in both groups undergoing radiotherapy and a minority treated with chemotherapy. We observed that the IS group had a meaningfully lower OS at 3, 5, and 10 years and lower DSS at 3, 5, and 10 years when compared with the non-IS group. We found a trend for a reduced recurrence-free survival in the IS group vs the non-IS group at 3 and 5 years that was not meaningful. The findings are consistent with single-institutional, multi-institutional, and meta-analysis studies reporting that immunosuppression is independently associated with inferior outcomes in cSCCHN despite similar, aggressive, multimodal treatment.5,6,10,11,12
The patterns of failure as defined by types of recurrences showed no significant difference between the 2 cohorts, including the risk of developing a second primary tumor or a local recurrence. Although the regional recurrence rate in the IS group was higher compared with the non-IS group and distant recurrences were found to be lower in the IS group, neither difference was meaningful. In this study, we observed no meaningful difference between certain high-risk pathologic features, including positive margins, peripheral nerve invasion, and lymphovascular invasion between the IS and non-IS cohorts.
Previous studies have shown that patients with immunosuppression and cSCC have a more aggressive behavior when compared with patients without immunosuppression.5,6,10 Previous studies have also compared the degree of differentiation of cSCCs among organ and stem cell transplant recipients (ie, patients with immunosuppression) and patients without immunosuppression and found no meaningful difference in the degree of tumor differentiation or aggressiveness between the 2 cohorts.13,14 However, a higher frequency of poorly differentiated cSCC has been reported with increased age independent of immune status.14,15 Patients with immunosuppression after organ and stem cell transplant present at a younger age than those without immunosuppression but not necessarily with more poorly differentiated tumors.14,15,16
On multivariate analysis, immunosuppression remained independently associated with DSS, with approximately twice the risk of death due to disease. Other factors associated with worse DSS were advanced age and locoregionally advanced disease. These findings are consistent with a meta-analysis10 concluding that immunosuppression in cSCCHN confers more than a 3-fold increased risk of death compared with no immunosuppression. Other studies have shown increased tumor and nodal categories as associated with poor outcomes.17
We further explored the association of the type of immunosuppression on survival. The most common form of immunosuppression we identified was type 1 or 2 diabetes treated with insulin, followed by lymphoma or leukemia, organ or stem cell transplant, autoimmune disease requiring treatment, and HIV. Five patients had immunosuppression in other forms, which included myelofibrosis, multiple myeloma, myelodysplasia, chronic immune demyelinating polyneuropathy, and lymphocytic meningitis. Subgroup analysis demonstrated that patients with a diagnosis of HIV or AIDS had the highest risk of death due to disease, followed by patients with hematopoietic malignant disease and organ and stem cell transplant. Diabetes was not independently associated with worse DSS. Previous studies have shown that patients with immunosuppression due to underlying hematologic malignant disease or organ and stem cell transplant were at a particularly high risk for poor disease control.10 However, we found no meaningful difference between cause of immunosuppression and DSS or OS on univariate analysis.
Because of the poor disease control demonstrated in patients with cSCCHN and immunosuppression, a need for novel treatment strategies remains. Unfortunately, clinical trials typically exclude patients with immunosuppression other than diabetes. The use of targeted therapies in the neoadjuvant setting, such as the epidermal growth factor receptor inhibitor gefitinib, can be well tolerated and did not interfere with definitive treatment in patients with aggressive cSCC.18 Cetuximab, a chimeric monoclonal antibody against epidermal growth factor receptor, resulted in a complete or a partial response in 21% to 28% of patients with cSCC and is a potential target for high-risk populations in particular,19,20 with 1 case report21 showing a complete resolution in an elderly patient with immunosuppression. Further clinical trials are warranted in high-risk populations for the use of epidermal growth factor receptor inhibitors.20
Checkpoint inhibitors have demonstrated great promise in the management of aggressive cSCC. Several small studies21,22,23,24,25,26 have shown meaningful activity using checkpoint inhibitors alone or in combination with standard therapies in advanced or metastatic cSCC. Based on a response rate of 50%, US Food and Drug Administration approval was granted for the use of cemiplimab, a PD-1 (programmed cell death 1) antagonist, in unresectable and metastatic cSCC.25 Further studies are needed to establish the safety and efficacy of targeted therapies and immune-modulating therapies, especially in the immunosuppressed population, because organ transplant recipients may have graft rejection with the use of these agents.27
Although the National Comprehensive Cancer Network recognizes immunosuppression as a high-risk feature of cSCC, current standard staging systems, including the eighth edition of the American Joint Committee on Cancer and Brigham and Women’s Hospital staging systems, fail to include immunosuppression status as a risk factor.28,29 Incorporating immunosuppression into staging may be beneficial in better risk-stratifying patients with cSCC. At a minimum, standardizing reporting of immunosuppression and the potentially underlying cause in patients with cSCC would allow for further population-based studies to define the role of immunosuppression in survival. We believe such studies will be important in understanding the role of novel treatment strategies for aggressive cSCC.
Strengths and Limitations
Limitations of this study include the retrospective design that relied on the accuracy of documentation with the possibility of important data not being available or accessible. Owing to the heterogeneous nature of the causes of immunosuppression in this cohort, we were unable to accurately measure and compare the degree or duration of immunosuppression. The subgroup analyses attempted to control for some of the variability in degree of immunosuppression. The risk of selection bias is also inherent. Although we attempted to adjust for these factors using multivariate analysis, not all nuances that influence treatment choice could be adequately accounted for in our models. In addition, this study was completed at a tertiary cancer center, potentially limiting the external validity of the findings.
Despite these limitations, this study demonstrates a large IS cohort of patients with cSCCHN. By including a variety of causes for immunosuppression, subgroup analysis was completed, improving our understanding of how different causes of immunosuppression affect DSS. This study also highlights the need for further studies in patients with cSCCHN and immunocompromise.
Conclusions
This retrospective cohort study found that immunosuppression was independently associated with outcome in patients with cSCCHN, with a 2.32-fold increased risk of disease-specific death after adjusting for age, history of skin cancer, recurrent or persistent disease status, disease stage, and treatment. On subgroup analysis, patients with HIV or AIDS had the worst prognosis in the IS group. Although surgery remains the mainstay treatment in these patients, these findings suggest that an improved understanding of adjuvant treatment options in these high-risk patients is required to improve their disease outcomes.
References
- 1.Kauvar AN, Arpey CJ, Hruza G, Olbricht SM, Bennett R, Mahmoud BH. Consensus for nonmelanoma skin cancer treatment, part II: squamous cell carcinoma, including a cost analysis of treatment methods [published correction appears in Dermatol Surg. 2016;42(3):443]. Dermatol Surg. 2015;41(11):1214-1240. doi: 10.1097/DSS.0000000000000478 [DOI] [PubMed] [Google Scholar]
- 2.Gray DT, Suman VJ, Su WP, Clay RP, Harmsen WS, Roenigk RK. Trends in the population-based incidence of squamous cell carcinoma of the skin first diagnosed between 1984 and 1992. Arch Dermatol. 1997;133(6):735-740. doi: 10.1001/archderm.1997.03890420073008 [DOI] [PubMed] [Google Scholar]
- 3.Pritchett EN, Doyle A, Shaver CM, et al. Nonmelanoma skin cancer in nonwhite organ transplant recipients. JAMA Dermatol. 2016;152(12):1348-1353. doi: 10.1001/jamadermatol.2016.3328 [DOI] [PubMed] [Google Scholar]
- 4.Ulrich C, Arnold R, Frei U, Hetzer R, Neuhaus P, Stockfleth E. Skin changes following organ transplantation: an interdisciplinary challenge. Dtsch Arztebl Int. 2014;111(11):188-194. doi: 10.3238/arztebl.2014.0188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Manyam BV, Garsa AA, Chin RI, et al. A multi-institutional comparison of outcomes of immunosuppressed and immunocompetent patients treated with surgery and radiation therapy for cutaneous squamous cell carcinoma of the head and neck. Cancer. 2017;123(11):2054-2060. doi: 10.1002/cncr.30601 [DOI] [PubMed] [Google Scholar]
- 6.Manyam BV, Gastman B, Zhang AY, et al. Inferior outcomes in immunosuppressed patients with high-risk cutaneous squamous cell carcinoma of the head and neck treated with surgery and radiation therapy. J Am Acad Dermatol. 2015;73(2):221-227. doi: 10.1016/j.jaad.2015.04.037 [DOI] [PubMed] [Google Scholar]
- 7.Smith KJ, Hamza S, Skelton H. Histologic features in primary cutaneous squamous cell carcinomas in immunocompromised patients focusing on organ transplant patients. Dermatol Surg. 2004;30(4, pt 2):634-641. doi: 10.1097/00042728-200404020-00011 [DOI] [PubMed] [Google Scholar]
- 8.Marrazzo G, Thorpe R, Condie D, Pinho MC, Srivastava D. Clinical and pathologic factors predictive of positive radiologic findings in high-risk cutaneous squamous cell carcinoma. Dermatol Surg. 2015;41(12):1405-1410. doi: 10.1097/DSS.0000000000000526 [DOI] [PubMed] [Google Scholar]
- 9.Koyfman SA, Cooper JS, Beitler JJ, et al. ACR Appropriateness Criteria® aggressive nonmelanomatous skin cancer of the head and neck. Head Neck. 2016;38(2):175-182. doi: 10.1002/hed.24171 [DOI] [PubMed] [Google Scholar]
- 10.Elghouche AN, Pflum ZE, Schmalbach CE. Immunosuppression impact on head and neck cutaneous squamous cell carcinoma: a systematic review with meta-analysis. Otolaryngol Head Neck Surg. 2019;160(3):439-446. doi: 10.1177/0194599818808511 [DOI] [PubMed] [Google Scholar]
- 11.Lam JKS, Sundaresan P, Gebski V, Veness MJ. Immunocompromised patients with metastatic cutaneous nodal squamous cell carcinoma of the head and neck: poor outcome unrelated to the index lesion. Head Neck. 2018;40(5):985-992. doi: 10.1002/hed.25069 [DOI] [PubMed] [Google Scholar]
- 12.Thompson AK, Kelley BF, Prokop LJ, Murad MH, Baum CL. Risk factors for cutaneous squamous cell carcinoma recurrence, metastasis, and disease-specific death: a systematic review and meta-analysis. JAMA Dermatol. 2016;152(4):419-428. doi: 10.1001/jamadermatol.2015.4994 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Cheng JY, Li FY, Ko CJ, Colegio OR. Cutaneous squamous cell carcinomas in solid organ transplant recipients compared with immunocompetent patients. JAMA Dermatol. 2018;154(1):60-66. doi: 10.1001/jamadermatol.2017.4506 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Stenman C, Gonzalez H, Gillstedt M, et al. Degree of differentiation of cutaneous squamous cell carcinoma: a comparison between a Swedish cohort of organ transplant recipients and immunocompetent patients. Dermatol Pract Concept. 2018;8(4):330-336. doi: 10.5826/dpc.0804a18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Harwood CA, Proby CM, McGregor JM, Sheaff MT, Leigh IM, Cerio R. Clinicopathologic features of skin cancer in organ transplant recipients: a retrospective case-control series. J Am Acad Dermatol. 2006;54(2):290-300. doi: 10.1016/j.jaad.2005.10.049 [DOI] [PubMed] [Google Scholar]
- 16.Lindelöf B, Dal H, Wolk K, Malmborg N. Cutaneous squamous cell carcinoma in organ transplant recipients: a study of the Swedish cohort with regard to tumor site. Arch Dermatol. 2005;141(4):447-451. doi: 10.1001/archderm.141.4.447 [DOI] [PubMed] [Google Scholar]
- 17.Blechman AB, Carucci JA, Stevenson ML. Stratification of poor outcomes for cutaneous squamous cell carcinoma in immunosuppressed patients using the American Joint Committee on Cancer Eighth Edition and Brigham and Women's Hospital staging systems. Dermatol Surg. 2019;45(9):1117-1124. doi: 10.1097/DSS.0000000000001774 [DOI] [PubMed] [Google Scholar]
- 18.Lewis CM, Glisson BS, Feng L, et al. A phase II study of gefitinib for aggressive cutaneous squamous cell carcinoma of the head and neck. Clin Cancer Res. 2012;18(5):1435-1446. doi: 10.1158/1078-0432.CCR-11-1951 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dereure O, Missan H, Girard C, Costes V, Guillot B. Efficacy and tolerance of cetuximab alone or combined with chemotherapy in locally advanced or metastatic cutaneous squamous cell carcinoma: an open study of 14 patients. Dermatology. 2016;232(6):721-730. doi: 10.1159/000461578 [DOI] [PubMed] [Google Scholar]
- 20.Maubec E, Petrow P, Scheer-Senyarich I, et al. Phase II study of cetuximab as first-line single-drug therapy in patients with unresectable squamous cell carcinoma of the skin. J Clin Oncol. 2011;29(25):3419-3426. doi: 10.1200/JCO.2010.34.1735 [DOI] [PubMed] [Google Scholar]
- 21.Chang ALS, Kim J, Luciano R, Sullivan-Chang L, Colevas AD. A case report of unresectable cutaneous squamous cell carcinoma responsive to pembrolizumab, a programmed cell death protein 1 inhibitor. JAMA Dermatol. 2016;152(1):106-108. doi: 10.1001/jamadermatol.2015.2705 [DOI] [PubMed] [Google Scholar]
- 22.Chen A, Ali N, Boasberg P, Ho AS. Clinical remission of cutaneous squamous cell carcinoma of the auricle with cetuximab and nivolumab. J Clin Med. 2018;7(1):E10. doi: 10.3390/jcm7010010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Falchook GS, Leidner R, Stankevich E, et al. Responses of metastatic basal cell and cutaneous squamous cell carcinomas to anti-PD1 monoclonal antibody REGN2810. J Immunother Cancer. 2016;4:70. doi: 10.1186/s40425-016-0176-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ferris RL, Blumenschein G Jr, Fayette J, et al. Nivolumab for recurrent squamous-cell carcinoma of the head and neck. N Engl J Med. 2016;375(19):1856-1867. doi: 10.1056/NEJMoa1602252 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Migden MR, Rischin D, Schmults CD, et al. PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma. N Engl J Med. 2018;379(4):341-351. doi: 10.1056/NEJMoa1805131 [DOI] [PubMed] [Google Scholar]
- 26.Seiwert TY, Burtness B, Mehra R, et al. Safety and clinical activity of pembrolizumab for treatment of recurrent or metastatic squamous cell carcinoma of the head and neck (KEYNOTE-012): an open-label, multicentre, phase 1b trial. Lancet Oncol. 2016;17(7):956-965. doi: 10.1016/S1470-2045(16)30066-3 [DOI] [PubMed] [Google Scholar]
- 27.Lipson EJ, Bagnasco SM, Moore J Jr, et al. Tumor regression and allograft rejection after administration of anti–PD-1. N Engl J Med. 2016;374(9):896-898. doi: 10.1056/NEJMc1509268 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Amin MD, Edge SB, Greene FL, et al. , eds. AJCC Cancer Staging Manual. 8th ed New York, NY: Springer International Publishing; 2017. doi: 10.1007/978-3-319-40618-3 [DOI] [Google Scholar]
- 29.Jambusaria-Pahlajani A, Miller CJ, Quon H, Smith N, Klein RQ, Schmults CD. Surgical monotherapy versus surgery plus adjuvant radiotherapy in high-risk cutaneous squamous cell carcinoma: a systematic review of outcomes. Dermatol Surg. 2009;35(4):574-585. doi: 10.1111/j.1524-4725.2009.01095.x [DOI] [PubMed] [Google Scholar]