The prognostic value of inositol polyphosphate 5‐phosphatase in cutaneous squamous cell carcinoma in the general population

Abstract

The prognostic value of inositol polyphosphate 5-phosphatase in cutaneous squamous cell carcinoma in the general population Dear Editor, Cutaneous squamous cell carcinoma (cSCC) is one of the most common malignancies, affecting over 700,000 individuals annually. Inositol polyphosphate 5-phosphatase A (INPP5A) is a cell membrane protein that negatively regulates inositol signaling and may be critical in cell signaling. Our prior research demonstrated loss of INPP5A expression was significantly predictive of poor overall survival in high-risk, recurrent, and metastatic cSCC. The prognostic value of INPP5A in the general population with cSCC is unknown. We therefore investigated the correlation of INPP5A expression in cSCC with staging and outcomes among a cohort of patients representative of the general population. Patients were randomly selected with stratification on clinic site, sex, race, ethnicity, and overall death to be representative of the entire Mayo Clinic cSCC population. Clinical data were obtained through chart review. All cases underwent histopathologic rereview and staging (using American Joint Committee on Cancer [AJCC] 7 and 8 editions and Brigham and Women’s Hospital [BWH] staging systems) by dermatopathologists (D.J.D, S.A.N) who were blinded to patient outcomes. Samples were stained with INPP5A, and expression levels were scored on a 4-point scale (0-3). A score of 3 indicated normal expression, 2 indicated partially diminished expression, 1 indicated significantly diminished expression, and 0 indicated total loss of expression, as compared to normal matched skin. High staining expression (HSE) was defined as a score of 2 or 3 and low staining expression (LSE) as a score of 0 or 1. We identified 454 primary cSCC cases comprised of 75 in situ and 379 invasive (Clark level > I) tumors. The majority (84%) of tumors had HSE of INPP5A. Subanalysis was performed on invasive tumors. Tumor characteristics for the overall cohort and invasive subset as well as univariate and multivariate Cox regression are detailed in Table 1. Our invasive subpopulation had 86% T1 tumors, 13% T2a tumors, and 1% T2b tumors by BWH staging. At 4 years of follow-up, few patients had poor outcomes: six (1.3%) local recurrence, no nodal metastasis, four patients (0.8%) died of unknown causes (disease-specific death cannot be ruled out), and 30% overall deaths. Although our cohort had few poor outcomes, we found a trend toward decreased overall survival with LSE of INPP5A on univariate analysis (hazard ratio 1.51; standard deviation 0.94, 2.4; P = 0.085). Increasing loss of INPP5A expression has been observed in the progression of actinic keratoses to welldifferentiated cSCC to poorly-differentiated metastatic cSCC, indicating its potential role in the development of carcinogenesis. Prior research has demonstrated the loss of INPP5A in high-risk, recurrent, and metastatic cSCC and oropharyngeal squamous cell carcinoma. Our present study failed to corroborate these findings in a cohort of patients representative of the general population with cSCC, possibly due to the lack of poor outcomes in the overall cohort underpowering our dataset. Our data do not support utilization of INPP5A for risk stratification of cSCC among the general population.