Conference Coverage

Study evaluates number of needed to refer, biopsy for diagnosing a melanoma



The number of suspicious lesions needed to refer to diagnose melanoma was 31.5, while the number needed to biopsy was 7.5. At the same time, the number needed to refer to diagnose non-melanoma skin cancer was 4 and the number needed to biopsy was 1.5.

Dr. Nikolai Klebanov, department of dermatology, Massachusetts General Hospital, Boston.

Dr. Nikolai Klebanov

The findings come from a retrospective review of 707 patients referred to a tertiary medical center dermatology practice for suspicious lesions, presented in a poster session at the virtual annual meeting of the American Academy of Dermatology

“Multiple studies in the dermatology literature have looked at the number needed to treat (NNT) as a quality metric for dermatology clinics, where a lower number is ‘better,’” the study’s first author, Nikolai Klebanov, MD, said in an interview following the virtual meeting. “Our particular study is unique in that we estimated both the number needed to refer and number needed to biopsy to closely examine the process of referrals for suspicious lesions from primary care settings to specialists. We also looked closely at the underlying patient-centered characteristics, which could be used by all clinicians to streamline the referral process by reducing the volume of low-risk referrals.”

Dr. Klebanov, of the department of dermatology at Massachusetts General Hospital, Boston, and his associates reviewed 707 unique patient visits to the department during July 2015–February 2016. They calculated the number needed to refer and biopsy for melanoma as the ratio of biopsy-proven melanoma diagnoses among benign and dysplastic nevi and seborrheic keratoses. For nonmelanoma skin cancer, they used the ratio of basal and squamous cell carcinoma among actinic keratoses and seborrheic keratoses.

Of the 707 patients, 54% were female, and males were slightly older than females (a mean of 58 vs. 54 years, respectively). The researchers found that lesions were more commonly benign among all age groups, while the frequency of premalignant and malignant lesions such as actinic keratoses, nonmelanoma skin cancer, and melanoma was highest for males and increased with age. Nevi were the most common benign diagnosis among patients 39 years of age and younger, while seborrheic keratoses were more common among patients aged 40 years and older.

The researchers found that the number needed to treat for melanoma was 31.5 and the number needed to biopsy was 7.5, which represents a 4.2-fold difference. Meanwhile, the number needed to refer for nonmelanoma skin cancer was 4, and the number needed to biopsy was 1.5, which represents a 2.7-fold difference. Despite variable rates of skin cancer between demographics, the biopsy rate ranged between 18% and 30%, for a mean of 23.4%.

“We found that most young patients referred for a ‘suspicious lesion’ on clinical prebiopsy assessment by the dermatologist were determined to actually have a benign nevus, and that older patients were most likely to have a seborrheic keratosis as the underlying lesion,” Dr. Klebanov said. “Among the minority of patients in each demographic group who were selected for biopsy, those lesions which were found to be benign were also largely nevi and keratoses. Even by being mindful of just the patient’s age, primary care providers can follow patients clinically with a tailored differential diagnosis in mind before referral, and dermatologists can reduce the number of biopsies they perform on patients who are being referred.”

He added that he and his colleagues were surprised that despite very low rates of skin cancer in young patients, and thus different pretest probabilities of cancer, biopsy rates across demographics were consistently around 20%. “We also found a disproportionate number of female patients younger than age 40 who were referred for suspicious lesions, while in the older age groups, the ratio of males to females was approximately equal.”

Dr. Klebanov acknowledged certain limitations of the study, including its single-center, retrospective design, and that information was not collected on patients’ family history of skin cancer, Fitzpatrick skin type, nor the clinical course of the lesion while it was followed by the primary care office. “The nuanced differences in these factors may certainly play a role in decisions for individual patients,” he said.

The study’s principal investigator was Hensin Tsao MD, PhD, clinical director of the MGH Melanoma & Pigmented Lesion Center The work was supported by the Alpha Omega Alpha Carolyn Kuckein Research Fellowship. The researchers reported having no financial disclosures.

SOURCE: Klebanov N et al. AAD 20. Abstract 15881.

Next Article: