Новый метод диагностики рака кожи сокращает необходимость в инвазивных биопсиях

Handheld reflectance confocal microscopy (HH-RCM) significantly reduces the need for invasive biopsies compared with dermoscopy alone for the diagnosis of head and neck basal cell carcinoma (BCC), according to study results published in the Journal of the European Academy of Dermatology and Venereology. Authors of a prospective study assessed the diagnostic accuracy of dermoscopy combined with HH-RCM for diagnosing and subtyping head and neck BCC as well as the potential of this combined approach for reducing diagnostic biopsies. The study was conducted at a single center in the Netherlands from 2017 to 2022 and enrolled adults with suspected primary head and neck BCC as assessed by dermatologists’ naked eye. Lesions were assessed with both dermoscopy and HH-RCM (DS+HH-RCM), and diagnostic confidence was based on a 3-point Likert scale. All histologic specimens were evaluated by 2 expert blinded pathologists, with discrepancies resolved by a separate pathologist. The diagnostic performance of DS+HH-RCM was assessed according to accuracy, sensitivity, specificity, and positive/negative predictive value. The analysis included 340 lesions from 270 patients, of which 286 (84%) were biopsy-proven BCC, with a 16% false-positive rate for naked eye examination. The patients had a median age at diagnosis of 72 years, 53% were women, and 64% had Fitzpatrick skin type II. Although HH-RCM has increased diagnostic accuracy compared to dermoscopy, it cannot completely replace diagnostic punch biopsies when management relies solely on BCC subtyping. The high diagnostic confidence rate was 33% for naked eye assessment compared with 49% (P <.001) and 88% (P =.001) for blinded dermoscopy and HH-RCM expert evaluations, respectively. The DS+HH-RCM BCC approach had high confidence of 83% (n=281; P <.001). The proportion of BCC subtyping that had high diagnostic confidence increased for dermoscopy (66%; n=180) compared with naked eye evaluation (42%; n=114; P =.001). Both HH-RCM (84%; n=229) and DS+HH-RCM diagnoses (68%; n=184) had an increased proportion of high-confidence subtyping compared with naked eye evaluation, although the difference was not significant compared with dermoscopy alone (P >.05). Consensus for the most aggressive histologic BCC subtype occurred in 89% (n=255) of biopsies and 88% (n=251) of excisions. The DS+HH-RCM approach correctly classified a significantly greater proportion of lesions compared with dermoscopy alone (P =.019), and the differences between DS+HH-RCM and HH-RCM and between dermoscopy and HH-RCM were not statistically significant (P >.05). Regarding overall BCC diagnosis, substantial agreement occurred between the real-time DS+HH-RCM assessment and blinded dermoscopy (κ =.620), with near perfect agreement between DS+HH-RCM and blinded HH-RCM (κ =.857). The observed, positive-, and negative-specific agreement for BCC diagnosis was 68% (95% CI, 58%-79%), 80% (95% CI, 71%-89%), and 24% (95% CI, 14%-34%) for dermoscopy. For HH-RCM, rates were 90% (95% CI, 84%-97%), 95% (95% CI, 89%-100%), and 61% (95% CI, 50%-72%), respectively. Outcomes for infiltrating BCC were 72% (95% CI, 61%-82%), 26% (95% CI, 16%-37%), and 82% (95% CI, 73%-91%) for dermoscopy and 78% (95% CI, 68%-88%), 64% (95% CI, 53%-75%), and 84% (95% CI, 76%-93%) for HH-RCM. Regarding high-confidence DS+HH-RCM diagnostic outcomes, the sensitivity and specificity for a BCC diagnosis were 98.8% (95% CI, 96.5%-99.7%) and 94.1% (95% CI, 80.3%-99.3%), respectively. Non-superficial BCC subtyping had similar sensitivity (92.5% vs 95.5%) and an increase in specificity from 44.8% (95% CI, 26.7%-62.9%) to 66.7% (95% CI, 38.4%-88.2%). For infiltrating BCC, a decrease was observed in sensitivity (52.3% vs 58.8%), with an increase in specificity from 79.6% (95% CI, 74.3%-84.9%) to 89.0% (95% CI, 82.7%-93.6%). Limitations of the study include the single-center design, and use of a single bedside reader for dermoscopy and HH-RCM may have overestimated the diagnostic performance. Also, a direct comparison between punch biopsy and excision outcomes is limited by incorporation bias, and the study did not correct for biopsy quality and depth. “Although HH-RCM has increased diagnostic accuracy compared to dermoscopy, it cannot completely replace diagnostic punch biopsies when management relies solely on BCC subtyping,” the study authors concluded. This article originally appeared on Dermatology Advisor