LAUSR

We read with interest the article by Turner et al., on the long-term imaging surveillance of patients with resected stages 3A to 3D melanoma. Based on the data presented, together with existing literature, we came to different conclusions from those of the authors. We concluded that surveillance imaging should not be routinely used for patients with stage 3 disease. Beginning with the abstract, a typographic error mistakenly states that the falseto true-positive ratio was 1:2. Conversely, the text of the article states that the ratio was 2:1 at best during the entire follow-up period and worsened to 4:1 during later tests. Indeed, the authors recognize a limitation to scheduled computed tomography (CT) or positron emission tomography (PET)/CT scans because ‘‘there is likely a point at which false-positives outnumber true-positives.’’ However, Table 2 demonstrates that falsepositive tests outnumbered true-positive tests at all time points. The detrimental impact of such a high false-positive rate is acknowledged, with 46 % of 332 patients receiving ‘‘684 further investigations, procedures, clinic visits, and referrals.’’ For those affected by a false-positive test, this was an average of 4.5 interventions per patient. Furthermore, numerous patients were affected by a total of 34 invasive procedures for benign lesions. This fundamental harm to patients did not include the financial burden or radiation dosage associated with the 1847 scans performed for correct identification of 79 recurrences. The reported rate is equivalent to 7 years of background radiation, scheduled every 6 to 12 months, per patient. The paucity of clinical utility in baseline imaging has been established by other studies. Similarly, Ravichandran, et al., concluded that ‘‘obtaining a baseline PET-CT within 3 months of initial diagnosis is unlikely to change initial management.’’ Although they viewed the initial scan’s value ‘‘as a baseline study for whom subsequent surveillance is planned,’’ they failed to give evidence concerning the benefit of subsequent scans other than a diagnosis of subclinical recurrence. Furthermore, among those with high-risk stages 3B to 3C disease and in-transit metastases (ITM), Holtkamp, et al., demonstrated that PET-CT failed to identify known ITM accurately for 47 % of the patients. They found that PET-CT changed the management for 4 of 25 patients, including total lymph node dissection (TLND) for two patients. Even for these high-risk patients, the findings showed no evidence of survival benefit with TLND in subclinical disease. Although 75 % of the patients in the Turner et al., study had stages 3B to 3D disease and may have undergone imaging to measure response to systemic therapy, the only benefit for the stage 3A patients was a diagnosis of subclinical disease recurrence. Again, the findings showed no significant evidence of a survival benefit in this group of patients for whom no approved systemic treatment exists. In the end, the question remains as to the benefit of a radiologic diagnosis of asymptomatic nodal or distant disease. With the results of the Multicenter Selective Lymphadenectomy Trial (MSLT)-1 and MSLT-2, we know that the diagnosis and surgical treatment of clinically undetectable nodal disease provides no survival benefit for Society of Surgical Oncology 2021