Early time to recurrence predicts worse survival in patients with localized or regionally advanced cutaneous melanoma

Abstract To investigate the prognostic significance of time to recurrence (TTR) for overall survival (OS) and survival after recurrence (SAR) in patients with localized or regionally advanced cutaneous melanoma. A total of 731 cutaneous melanoma patients with an initial diagnosis of 8th American Joint Committee on Cancer (AJCC) clinical stage I‐III were included in this study. The prognostic factors associated with OS and SAR were estimated through Kaplan‐Meier and Cox regression analysis. Of the total cohort, 329 patients (45%) died, and 418 patients (57%) experienced recurrence. The median follow‐up and TTR were 55.6 months and 9.6 months, respectively. A total of 141 patients (19%) experienced recurrence in <6 months, and 277 patients (38%) experienced recurrence in ≥6 months. Patients with stage III and positive lymph node dissection (LND) were more common in the early TTR group than in the late TTR group. Both the OS and SAR rates at 5 years and 10 years in the early TTR group were significantly poorer than those in the late TTR group (P < .001 and P = .008, respectively). Furthermore, early TTR, along with truncal tumor, higher TNM stage and therapeutic variables (extended resection, LND and adjuvant therapy), were significant independent predictors of worse OS and SAR in multivariate analysis (all P < .05). Early TTR predicts worse survival and could be considered an independent prognostic factor for patients with localized or regionally advanced cutaneous melanoma. TTR should be evaluated in all patients with recurrence to guide post‐recurrence risk stratification and follow‐up schedules.


| Data collection and treatment
The clinical and pathological characteristics of all patients were collected from medical records. The results from sentinel lymph node biopsy (SLNB) were collected from clinical notes and pathological test reports. Tumor stage was assessed according to the AJCC 8th edition melanoma staging system. 5 According to the published guidelines, 4 extended excision was defined as the primary tumor resection with a safety margin of 1 cm for Breslow thickness below 2 mm and 2 cm for Breslow thickness above 2 mm. SLNB was indicated in patients with a tumor thickness of >1 mm or > 0.75 mm accompanied with additional risk factors such as mitotic rate or ulceration. Regional lymph node dissection (LND) was recommended for patients with positive SLNB or stage III disease diagnosed by physical examination, imaging or pathology.
Adjuvant therapy refers to those supplementary treatments except surgery, which mainly consists of adjuvant radiotherapy, highdose interferon α-2b and cytotoxic drugs such as dacarbazine, temozolomide or other chemotherapy drugs. Generally, patients with high-risk (stage IIB IIIA) and very high-risk (stage IIIBIIIC) melanoma were recommended to receive adjuvant therapy. Adjuvant radiotherapy was only considered in cases of inadequate surgical resection or bulky disease.

| Follow-up
An independent follow-up department in our hospital performed regular follow-up. The latest follow-up date of this study (October first, 2019) or death was considered as the final survival follow-up time.
Overall survival (OS) was defined as the time from primary tumor resection to any cause death or the last follow-up. Recurrence-free survival (RFS) was defined as the time from the resection to the first recurrence or the last follow-up. SAR was calculated from the date of the first recurrence until the date of death or the last follow-up. For recurrent patients, we evaluated the type of first recurrence, duration of TTR, presence of distant metastases (either as first or subsequent recurrence), and SAR. In accordance with that in previous studies, 7,13 the type of first recurrence was classified as local recurrence (within 5 cm of the primary scar), intralymphatic recurrence, regional lymph node recurrence and distant recurrence (distant organs or lymph nodes).

| RESULTS
The characteristics of the 731 patients are shown in Table 1    SAR is an optimal study endpoint for cutaneous melanoma, which is characterized by a relatively high rate of recurrence after initial treatment. However, compared with OS, there is much less work exclusively assessing the predictors of SAR. In our study, along with TTR, tumor site, TNM stage and therapeutic variables (extended resection, LND and adjuvant therapy) remained the most crucial predictors of both OS and SAR. This result corroborated with those of previous reports, 8,9,18,19 which indicated that the biological invasiveness of the primary tumor might be maintained in the subsequent local recurrence and confirmed the prognostic importance of the radicality of initial treatment.
Long-term follow-up is inevitable for the analysis of time to first recurrence and SAR. The median follow-up time of our study was 55.6 months (IQR: 33.9 94.2 months). Usually, the recommended follow-up for melanoma is 5 to 10 years. In addition, the median TTR of recurrent patients was 9.6 months (IQR: 4.8 20.7 months), which was shorter than that in other reports 17  Their results indicated that the prognosis of Chinese patients with malignant melanoma was suboptimal. However, it is noteworthy that their cohort was mixed with mucosal melanoma and other subtypes.
Overall, there is large variability in survival and recurrence among different counties and regions.
In addition, we investigated risk factors for RFS. The results showed that large size, positive surgical margin and higher TNM stage were significantly associated with poor RFS. These findings were consistent with those of previous studies. 7 It is noteworthy that adjuvant therapy was significantly associated with worse survival and higher recurrence. This phenomenon might be attributed to the higher proportion of patients with advanced stages and unfavorable tumor characteristics in the adjuvant therapy group. Moreover, variables such as sex, tumor site, tumor thickness, ulceration, SLNB and extended resection were significantly associated with RFS in the univariate analysis. Therefore, male patients with truncal tumors without extended resection should probably be given more attention in long-term follow-up.
Finally, there are some limitations to be noted. First, this was a retrospective study with the accompanying inherent sources of bias.
Second, the enrolled patients received treatment over a spanning period of over 20 years, during which great progress in treating melanoma had occurred. Third, all patients enrolled in this study were selected from one hospital, which means our conclusions should be verified in a larger population from multiple centers. In addition, some clinicopathological variables were not incorporated into the analysis, such as mitotic rate and histologic subtypes. These data were not obligatory in previous pathological reports, consequently leading to missing data in many patients.

| CONCLUSION
Early TTR predicts worse survival and could be considered an independent prognostic factor for patients with localized or regionally advanced cutaneous melanoma. TTR should be evaluated in all patients with recurrence to guide post-recurrence risk stratification and follow-up schedules.