About This Tool

The CTCL Risk Assessment Tool was developed to help dermatologists identify patients who may be at higher risk for Cutaneous T-Cell Lymphoma (CTCL). Early diagnosis of CTCL can be challenging due to its resemblance to common skin conditions like eczema and psoriasis, often leading to delays in diagnosis and treatment.

This tool uses a set of clinical indicators that have been associated with CTCL to generate a risk assessment. It is designed to be used by general dermatologists as an aid in clinical decision-making, particularly when considering whether to perform a biopsy or refer to a specialist.

The risk calculator is based on clinical research and expert consensus on the most common presenting features of CTCL. It is intended to supplement, not replace, clinical judgment and does not provide a definitive diagnosis.

The risk assessment is a transparent weighted score derived from a study in the Geskin lab, which trained a supervised machine-learning classifier (Random Forest) on structured clinical data extracted from electronic health records prior to diagnosis. That study used key binary historical variables—including history of multiple biopsies, failed topical steroids, other failed therapies, scaly patches or plaques, erythema, xerosis, pruritus, and presence of other rashes—to distinguish patients with cutaneous T-cell lymphoma (CTCL) from those with benign inflammatory skin conditions. Prioritizing sensitivity to minimize false negatives, that research model reported approximately 72% accuracy, 87% sensitivity, and 55% specificity in its development cohort. This calculator is a simplified weighted score based on those relative feature weightings; the reported figures reflect the original research model and have not been prospectively validated for this tool.

It's important to note that this tool provides a risk assessment, not a diagnosis. A high-risk result should prompt consideration of appropriate next steps, which may include biopsy, referral to a specialist, or additional testing.

Users should be aware of the following limitations:

• This tool is designed for use by healthcare professionals and is not intended for self-diagnosis by patients
• The risk assessment is based on clinical features and does not incorporate histopathological, immunophenotypic, or molecular data
• A low-risk result does not rule out CTCL, particularly in cases with strong clinical suspicion
• The tool has not been validated in all populations and may perform differently in various demographic groups
• Clinical judgment should always take precedence over the risk assessment provided by this tool