Desmoplastic Melanoma Has High Local Recurrence Rate After Wide Local Excision

Desmoplastic melanoma (DM) has a local recurrence rate (LRR) of 21% after wide local excision (WLE), according to study findings published in Dermatologic Surgery.

Researchers conducted a systematic review and meta-analysis to assess LRR after excision of DM with WLE, Mohs micrographic surgery (MMS), or staged excision (SE), as well as tumor and treatment risk factors for local recurrence. They searched the PubMed, Embase, and Web of Science databases through December 13, 2021. Risk ratios (RR) were calculated using meta-analysis of binary outcome data.

A total of 55 studies met the eligibility criteria. There were 3 studies that reported LRR for DM excised with MMS, and 1 study that reported LRR for DM excised with SE. A total of 53 studies assessed WLE, including 51 that treated DM solely with WLE and 2 studies that treated DM with WLE or MMS. The 53 studies were published from 1971 to 2021 and included 3080 total DM cases.

The 4 studies that evaluated MMS or SE had 61 total DM cases. The small number and heterogeneity of the studies precluded calculating a statistically meaningful summary estimate of local recurrence.

Of the 53 studies that evaluated DM treated with WLE, most DMs occurred in men (66%; n=2020), and the weighted average mean age and median age were 63 years (n=1792) and 65 years (n=1249), respectively. The most common anatomic site was the head and neck (55%; n=1687).

The overall summary estimate of LRR for 2308 DMs treated with WLE from 47 studies that had nonduplicate data was 21% (95% CI, 0.16-0.28; I², 88%; P <.01). The follow-up was included for 1929 (84%) tumors with a weighted average mean of 51 months (n=660) and median of 65 months (n=1269).

There were 14 studies where LRR was reported to be associated with histologic excision margin status after WLE, and 9 studies reported DMs with negative margins or positive or unknown margins. Summary estimates of LRR by margin status in the 14 studies showed a lower LRR of 11% (95% CI, 0.07-0.17; I², 81%; P <.01) for DM excised with negative margins (n=1075) vs 49% (95% CI, 0.25-0.74; I², 72%; P =.03) for DM with positive or unknown margins (n=91; P <.01 for subgroup differences).

In the 9 studies that assessed DMs excised with negative margins (n=688) and with positive or unknown margins (n=91), positive or unknown margins were associated with a 4.39-fold increased LRR (RR, 4.39; 95% CI, 2.45-7.86; P <.01 for test for overall effect; I², 60%; P =.01).

Analysis of 13 studies that included neurotropic (n=228) and nonneurotropic (n=416) DM found that neurotropism was associated with an increased risk for local recurrence (RR, 1.79; 95% CI, 1.34-2.38; P <.01 for test for overall effect; I², 7%; P =.38).

Data from 13 studies were used to calculate and compare summary estimates of LRR according to use of adjuvant radiation. The LRR was 7% (95% CI, 0.05-0.11; I², 0; P =1.00) for tumors treated with adjuvant radiation (n=287) compared with 21% (95% CI, 0.13-0.34; I², 85%; P <.01) for those treated without adjuvant radiation (n=572; P <.01 for subgroup differences).

Limitations of the analysis include the fact that studies were retrospective, single-institution studies with a high risk for bias. In addition, interstudy heterogeneity was significant, owing to different geographic locations, patient populations, and approaches to surgery.

“This systematic review and meta-analysis shows a high LRR after WLE of DM,” conclude the study authors. “Positive histological excision margins and neurotropism are associated with higher LRR. Additional studies are necessary to evaluate LRR after MMS or SE.”