Clinicopathological Characteristics Predicting Further Recurrence and Survival Following Resection of In-Transit Melanoma Metastases

Background In-transit metastases (ITMs) affect approximately 4% of patients with cutaneous melanoma. This study sought to identify clinical and pathological characteristics that predict further recurrence and survival following resection of ITMs. Patients and Methods Patients (n = 573) who underwent surgical resection of their first presentation of ITM following previous surgical treatment of an American Joint Committee on Cancer (AJCC) stage I–II melanoma between 1969 and 2017 were identified from an institutional database. Clinicopathological predictors of patterns of recurrence and survival following ITM resection were sought. Results The median time of ITM development was 2.4 years after primary melanoma resection. ITMs were most frequently located on the lower limb (51.0%). The most common melanoma subtype associated with ITM development was nodular melanoma (44.1%). After surgical resection of a first ITM, 65.4% of patients experienced recurrent disease. Most recurrences were locoregional (44.7%), with distant metastasis occurring in 23.9% of patients. Lower limb ITMs were more frequently associated with subsequent ITMs [odds ratio (OR) 2.41, p = 0.0002], and the lowest risk of distant metastasis (p < 0.0001) compared with other primary sites. Primary melanomas and ITM on head and neck, as well as the presence of ulceration, were associated with worse survival. Conclusions Recurrence after surgical resection of a first ITM was common. Patterns of recurrence differed according to anatomical site; further ITM recurrences were more likely for lower limb ITMs, which were also associated with longer distant recurrence-free survival. Distant metastasis was more common for ITM on the head and neck, with worse survival. Supplementary Information The online version contains supplementary material available at 10.1245/s10434-022-11997-0.

In-transit metastases (ITMs) are cutaneous or subcutaneous metastases located between the primary tumor site and the regional node field, [1][2][3][4] and are thought to occur because of intralymphatic (or possibly angiotropic) tumor spread. 5 ITMs occur in approximately 4% of all patients with invasive cutaneous melanomas, rising to 11% in patients with thicker primary melanomas. [5][6][7] Risk factors for developing ITMs include site of the primary melanoma (limb or trunk), primary tumor characteristics (increased Breslow thickness, higher mitotic rate, lymphovascular invasion, and the presence of ulceration), a positive sentinel lymph node biopsy, and macroscopic regional node involvement. 1,2,7,8 The latter pathological characteristics are also adverse prognostic indicators for patients with melanoma in general. 5,7,9 Surgical resection remains the standard treatment for operable ITMs. However, there are numerous other locoregional treatment modalities available for ITMs, ranging from topical therapies to isolated limb infusion. 1,3,4 Furthermore, adjuvant systemic treatment is now often recommended for patients with resected stage III melanoma, which includes ITMs. 5 This includes immunotherapy with checkpoint inhibitors (such as nivolumab, pembrolizumab, or ipilimumab), or targeted therapy in patients with BRAF-mutant melanoma (such as combination dabrafenib and trametinib). 10,11 Recent trials suggest these drugs are effective treatment after resection of ITMs, however patients with ITMs represent only a very small proportion of the studied cohorts 11,12 or were excluded from these trials. 10 The use of neoadjuvant systemic therapies for stage III melanoma is being assessed in ongoing clinical trials. [13][14][15] The use of drug therapies continues to expand in both the adjuvant and neoadjuvant settings. In the absence of other biomarkers, which may become available with time, it is important to identify high-and low-risk subpopulations of patients with ITMs from presently available data to guide management decisions and better stratify patients entering clinical trials. This stratification of patients may spare low-risk patients from adverse events associated with systemic therapy when they are less likely to benefit from treatment, and identify patients at high risk of recurrence, particularly distant recurrence, who stand to gain most from systemic therapy. Hence, this study sought to describe the patterns of recurrence following surgical excision of a first ITM(s) and evaluate the clinical and pathological characteristics that predict recurrence and survival following ITM resection.

Patient Population
Patients who underwent resection for a first presentation of isolated ITM following previous treatment of a primary AJCC stage I or II cutaneous melanoma at Melanoma Institute Australia (MIA), Sydney, between 1969 and 2017, were identified from a prospectively maintained database. In this database, ITMs are defined as cutaneous or subcutaneous metastases in the same part of the body, separated from the primary lesion by greater than 5 cm, but not in the draining node field. Patients who underwent initial resection of a primary cutaneous melanoma at another center and were referred to MIA for management of their ITM, were included if there was sufficient information on primary tumor pathological characteristics and follow-up.
Patients who first developed other regional or systemic recurrence and subsequently developed ITMs were excluded, as were those with synchronous lymph node or distant metastases at the time of initial ITM diagnosis. Nine patients who received adjuvant immunotherapy after ITM resection were also excluded. Retrieved information included age, gender, location of the primary tumor, melanoma subtype, Breslow thickness, ulceration, stage (AJCC 8th edition), number of ITMs, subsequent recurrence (date and site), date of death or last contact, and cause of death.

Management
All patients underwent surgery with the intention of achieving complete ITM excision with clear margins, as recommended by current Australian Melanoma Management Guidelines. 8 ITMs were considered operable where they were not only technically resectable but resection was felt to carry an acceptable morbidity profile compared with other treatment modalities. Sentinel node biopsy was not routinely performed for ITMs at MIA. Throughout the period of the study, clinical follow-up typically occurred every 3-4 months for the initial 2 years, then every 6 months to 5 years. Surveillance imaging and follow-up protocols for patients with ITMs treated at MIA changed over the period of the study. Surveillance ultrasound became more common through the 1990s. Since the early 2000s, patients have routinely undergone full-body computed tomography (CT), and more recently 18 Ffluorodeoxyglucose positron emission tomography/low dose CT (FDG-PET/CT) scans at the time of ITM diagnosis to assess for the presence of regional and distant metastases, with repeat full-body imaging at least every 12 months.

Statistical Analysis
Clinical and pathological parameters were analyzed using descriptive statistics, median (and range) for continuous variables and frequency (proportion) for categorical variables. Outcomes included disease-free survival (DFS), overall survival (OS), melanoma-specific survival (MSS), patterns of recurrence (including development of subsequent ITMs), and number of ITMs (count). Survival outcomes were calculated from the date of primary melanoma diagnosis and then from the date of resection of the first ITM. Survival curves were calculated using the Kaplan-Meier method, stratified by Breslow thickness and primary melanoma site. Survival differences between groups were assessed using the log-rank test. Multivariable Cox regression analyses were performed to identify prognostic factors associated with recurrence and survival outcomes in patients with resected ITMs. Associations between baseline factors and disease recurrence (versus no recurrence) were evaluated using univariable logistic regression.

Patient Population
Between 1969 and 2017, 1345 patients presented with first ITMs. Of these, 573 patients (43%) underwent surgical resection only and were included in the study. The other 772 ITM patients were found to have either multiple ITMs not suitable for complete resection, or concomitant nodal or distant disease; 174 patients had locoregional disease; 33 had distant metastases; and 565 had both locoregional and distant disease. Over the study period, 44,902 patients with melanoma were entered into the MIA database. Thus, patients presenting with isolated ITMs undergoing surgical resection represented 1.3% of all patients with melanoma who were seen and treated at the institution during the study period.
The median age of the 573 study patients was 68.4 years (range 14.4-95.5 years), and 321 (56.0%) were male. The most common primary melanoma subtypes of the patients presenting with first ITMs were nodular (n = 216, 44.1%) and superficial spreading (n = 188, 38.4%). The lower limb was the most common site for the primary melanoma (n = 292, 51.0%) ( Table 1). Median follow-up was 7 years after ITM resection. Median time from resection of the primary melanoma to diagnosis of ITM was 2.4 years (95% CI 2.08-2.75 years) with more than 90% occurring within 10 years of the primary melanoma diagnosis. A single ITM recurrence was the most common, with a range of 1-6 lesions ( Table 1).

Patterns and Timing of Recurrence Following ITM Resection
Following ITM resection, 375 patients (65.4%) experienced melanoma recurrence. Median time to recurrence (of any type) was 16 months from ITM resection (95% CI 13-18 months). The most common pattern of recurrence was locoregional (n = 256, 44.7%), including 122 patients (21.3%) with regional nodal disease and 94 patients (16.4%) recurring with further ITMs. There was a trend toward higher regional lymph node recurrence for ITMs resected from the lower limbs compared with those from primary sites on the head and neck, trunk, or upper limbs, although the difference was not statistically significant (HR 1.29, CI 0.78-2.15, p = 0.14).
Of the entire cohort, 137 patients (23.9%) developed a distant metastasis after ITM resection, either in isolation, or in combination with another recurrence type (Table 1; Fig. 1). Ten-year distant metastasis-free survival (DMFS) rate after ITM resection was 67.5% (95% CI 62.2-73.3%), with median DMFS not reached. Primary melanomas on the upper and lower limbs had significantly lower rates of distant metastasis compared with primary melanomas on trunk or the head and neck (HR 0.51 and HR 0.28, respectively, p \ 0.0001; Table 2). When patients were stratified by age, younger age (B 59 years) was associated with a higher rate of distant metastasis (p = 0.0142; Table 2). Breslow thickness and the presence of ulceration or lymphovascular invasion did not predict time to distant recurrence following initial ITM resection ( Table 2).

Further ITMs After ITM Resection
ITMs on the lower extremities were strongly associated with further ITMs after initial surgical resection (OR 2.41, 95% CI 1.35-4.31, p = 0.0002; Supplementary Table 2); this was not the case for other primary sites. Age, gender, Breslow thickness, melanoma subtype, ulceration, and lymphovascular invasion were not associated with further recurrence of ITMs after surgical resection. For patients who recurred with further ITMs after initial ITM resection, the median number of further (separate) ITM excisions ranged from 1 to 29 over the study period ( Table 1). Most of the further ITMs were within 2 years of initial ITM resection.

Patients Without Disease Recurrence After ITM Resection
Of the 573 patients who underwent surgical resection of ITMs 198 patients (34.6%) did not recur. These patients were older than those who experienced disease recurrence (median 72.5 versus 67.0 years, p = 0.0001). There was no difference between groups regarding gender, location of primary, Breslow thickness, melanoma subtype, presence of ulceration, AJCC stage at diagnosis, or number of ITMs at presentation.

Survival Following ITM Resection
From the time of primary melanoma diagnosis, the median melanoma-specific survival (MSS) and overall survival (OS), was 13 and 9.1 years, respectively. From the time of ITM resection, MSS was 6.4 years and OS was 4.1 years. The presence of ulceration in the primary melanoma was associated with worse MSS from both time of primary melanoma diagnosis (HR 1.49, p = 0.0099) and ITM  15% of data were missing for melanoma subtype and ulceration, and 44% of data missing for lymphovascular invasion resection (HR 1.52, p = 0.0038). From the time of primary diagnosis, melanomas on the head and neck, and those with increased Breslow thickness were associated with worse MSS (Table 3; Fig. 2). Primary site on the head and neck showed a trend toward worse MSS when calculated from time of ITM resection (p = 0.09; Table 3).
Regarding OS after ITM resection, age [ 77 years (HR 1.88, p = 0.0005), presence of ulceration in the primary melanoma (HR 1.38, p = 0.0079), and ITM on the head or neck (p = 0.0179) were associated with significantly worse OS (Supplementary Table 1).

DISCUSSION
Resection with clear margins remains the gold standard for management of operable melanoma ITMs. However, recurrence rates are high, and the timing, risk factors, and patterns of recurrence have not been well described in literature. Our study provides important information on favorable and unfavorable prognostic factors after ITM resection. This may assist in tailoring management, which today includes not only the options of a range of local and regional therapies, 1, 3, 4 but also the use of adjuvant and neoadjuvant systemic therapies. Our findings may also assist in the design of further trials involving patients with ITMs.

Melanoma Subtypes Associated with ITMs
Our finding that resected ITMs were more likely to occur after a nodular primary melanoma (44.1%) has not been reported previously. This suggests that nodular melanoma may have an inherently greater biological propensity for intralymphatic (or possibly angiotropic) tumor spread, which warrants further investigation.

Patterns and Timing of Recurrence After Resected ITM
We report a recurrence rate of 65.4% after resection of an initial ITM, in keeping with previous retrospective series suggesting recurrence rates (local, regional, or distant) of 63-72%. 16,17 Nearly half of the cohort experienced locoregional recurrence (44.7%), with similar rates of isolated regional lymph node and ITM recurrences. Less than a quarter developed distant metastases over the extended follow-up period. This should be borne in mind when considering management options in the contemporary era of systemic therapy for metastatic melanoma.

Risk Factors for Survival After ITM Resection
ITMs from primary melanomas on the head and neck were associated with the highest rates of distant metastasis and a trend to worse MSS from the time of ITM diagnosis. There is inherent difficulty in differentiating in-transit disease on the head and neck from systemic dermal metastases in the context of midline head and neck lesions, which may contribute to this finding, though it is considered unlikely to be a dominant confounder. Increased Breslow thickness was associated with worse MSS from both time of primary melanoma and from time of ITM despite not having a clear association with development of distant metastases.
Younger age was associated with high rates of distant metastasis after ITM resection but was only associated with worse MSS from time of primary melanoma, not from time

Patterns and Timing of Recurrence After ITM Resection
Following ITM resection, 16.4% of patients recurred with further ITMs, approximately four times greater than the rate of ITMs expected following excision of a primary cutaneous melanoma. This suggests an inherent biological propensity of the melanoma in these patients to spread and lodge intralymphatically. Patients with first ITMs on the lower limb were more than twice as likely to recur with further ITMs, compared with ITMs located elsewhere on the body. This may reflect increased intralymphatic hydrostatic pressure in the lower limb or possibly the greater surface area of the lower extremity upon which to recur.
The high rates of both nodal and distant recurrence after ITM resection support the use of regular imaging surveillance in this patient cohort. In particular combining ultrasound, owing to its high sensitivity and specificity in diagnosing nodal disease, 18 with PET/CT to assess both nodal and distant recurrences. 19 Ultimately, the disparate patterns of recurrence following resection of ITMs suggest the need for a multimodal approach to surveillance,

Use of Adjuvant Systemic Therapy After ITM Resection
In attempting to stratify high risk and low risk populations with respect to patterns of recurrence and survival, our results suggest that early consideration of systemic therapy in patients with head and neck ITMs may be advisable, given their increased risk of distant recurrence and worse survival. Similarly, ITMs in younger patients were associated with increased rates of subsequent distant recurrence, which appeared to impact MSS, suggesting that these patients stand to gain most from systemic therapy.
In contrast, patients with resected ITMs on the limbs had lower rates of distant metastasis, with improved MSS, compared with those with ITMs on the trunk or head and neck. This would appear to justify more aggressive local management of limb ITMs, particularly surgical resection when recurrences are amenable to operation, reserving systemic therapies for subsequent non-operable regional relapse or distant metastasis. Such an approach seeks to maximize potential benefit from systemic therapies in the context of ITMs while minimizing potential adverse effects.

Limitations
This study is strengthened by the large cohort size, extended follow-up time, and prospective nature of the database from which information was obtained. However, inherent limitations in the study design as an observational, single-institutional study remain. Due to the long study period , both the definition of operability and surveillance regimens undoubtedly changed over the course of the study. The concept of operable ITMs is not fixed, and is likely to have been a trend for less extensive surgery when other treatment modalities became available. Unfortunately, the MIA database lacks sufficient detail to reflect precisely the decision-making process with respect to operable versus nonoperable lesions in this cohort over the study period.
Surveillance practices also changed over the study period. The use of ultrasound increased through the 1990s, with improvements in technical sophistication and clinical expertise. Similarly, FDG-PET became incorporated into routine use a decade later. These changes likely improved staging and post-treatment surveillance, which would be expected to have improved outcomes by excluding those with occult metastatic disease from upfront surgery and allowing earlier detection and potential salvage at the time of relapse.
Despite the inherent limitations outlined above, this study documents the natural history of melanoma after resection of ITMs prior to the era in which potentially effective systemic therapies have become widely available to treat metastatic melanoma.   OPEN ACCESS This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.