Contemporary Approaches to In-Transit Melanoma

Abstract

In-transit melanoma represents a distinct disease pattern of heterogeneous superficial tumors. Many treatments have been developed specifically for this type of disease, including regional chemotherapy and a variety of directly injectable agents. Novel strategies include the intralesional delivery of oncolytic viruses and immunocytokines. The combination of intralesional or regional chemotherapy with systemic immune checkpoint inhibitors also is a promising approach. In the current review, we examine the general management of the workup of patients with in-transit disease, the range of available therapies, and recommendations for specific therapies for an individual patient.

INTRODUCTION

In-transit (IT) melanoma represents a distinct disease pattern whereby the disease recurs as dermal or subcutaneous nodules between the primary melanoma site and the regional lymph node basin. IT disease includes local recurrences within 2 cm of the primary lesion as well as multiple small lesions between the primary and regional nodal basin, which are frequently referred to as satellitosis. Currently, these lesions are all now more uniformly referred to as IT disease. IT disease can be heterogeneous in both the size and number of lesions, with most lesions ranging from 0.2 cm to 3 cm and the possibility that an individual patient may have one to more than 100 lesions.

After appropriate initial surgical therapy for primary melanoma, approximately 4% to 10% of patients develop IT disease at a median time of 18 months after primary excision.¹ IT metastases are considered to be American Joint Committee on Cancer (version 8) stage N1c (IIIB) without regional nodal involvement, stage N2c (IIIC) with one regional lymph node (LN) involved, and stage N3c (IIIC) with two or more regional LNs involved.² Risk factors for the development of IT disease include ulceration, lymphovascular invasion and Breslow depth of the primary tumor, the presence of LN involvement, age older than 50 years, and location on the lower extremity.³ Furthermore, 12% of first relapses after resection of stage II disease have been found to be local or IT, and 28% of relapses after resection of stage III disease were local or IT.^4,5

INITIAL WORKUP

On physical exam, IT melanoma can present in a variety of ways: lesions can be pigmented or nonpigmented; can appear as blisters, subcutaneous, or cutaneous nodules; and may even be mistaken as rashes. Any suspicious finding on physical exam, especially on the ipsilateral extremity or near the site of previously resected primary melanoma, should be sampled via fine-needle aspiration or punch biopsy to confirm the presence or absence of melanoma. If melanoma is confirmed, physicians should document the full extent of the disease, including the number and location of disease. Lesions that are ulcerated or that cause significant pain or discomfort should be noted. The physical examination should also focus on LN basins.

In addition to a medical history and physical examination to determine disease burden and functional status, the initial evaluation of patients with IT disease should also include the appropriate staging studies. Approximately 30% to 75% of patients who present with IT disease will have—or eventually develop—concurrent distant disease or regional nodal disease³; therefore, evaluation of a patient with biopsy-proven IT disease begins with a workup for distant disease that includes a total-body positron emission tomography–computed tomography and brain magnetic resonance imaging. Therapeutic options for patients who are positive for distant or nodal disease rely predominantly on systemic therapy. In patients who have a negative staging workup, therapeutic approaches will generally depend on multiple factors, including the patient’s overall functional status, disease burden, and disease location. An algorithm for a patient who initially presents with IT disease is shown in Figure 1, whereas Figure 2 depicts an algorithm for patients who have experienced failure with standard and first-line treatments.

Fig 1.

Algorithm for in-transit (IT) melanoma. Underlined therapies are approved therapies. (*) Therapies still under investigation. ILI, isolated limb infusion; HILP, hyperthermic isolated limb perfusion; HN, head and neck; LN, lymph node; SLNB, sentinel lymph node biopsy; T-VEC, talimogene laherparepvec.

Fig 2.

Algorithm for recurrent/treatment failure in-transit (IT) melanoma. ILI, isolated limb infusion; HILP, hyperthermic isolated limb perfusion; HN, head and neck; LN, lymph node.

EXCISION OF LIMITED IT DISEASE AND ROLE OF SENTINEL LYMPH NODE BIOPSY

Patients with no evidence of metastatic or nodal disease on initial workup are categorized as either resectable or nonresectable. A true definition of resectability should take into account any associated symptoms as well as whether the resection of individual lesions may provide a palliative benefit. Those patients with resectable disease should undergo complete resection, with or without adjuvant therapy and with or without sentinel LN (SLN) biopsy (SLNB).

For a patient who presents with a first IT recurrence that is resectable—generally fewer than three lesions and no one lesion greater than 5 cm in size—resection with or without SLNB can be performed. Advocates of SLNB of the IT lesion feel it should be strongly considered because, in addition to possibly changing systemic disease recommendations, it would change a patient’s prognosis. A multicenter study found that, in patients who were SLN negative at the time of primary melanoma excision (n = 44), 12 (27%) of 44 patients had a positive SLN at the time the local recurrence or IT melanoma was mapped.⁶ In a large retrospective study (N = 380), patients with IT disease and nodal metastases had a 36% 5-year survival compared with a 60% 5-year survival for those with IT disease and no nodal metastases, suggesting that LN involvement is an important prognostic factor for patients with IT disease.⁷ However, the lack of survival data supporting the removal of microscopic nodal disease in patients with stage III disease makes it important to engage a multidisciplinary tumor board before performing this procedure.

ADJUVANT THERAPY AFTER COMPLETE RESECTION

Following resection, adjuvant treatment options include systemic therapies, such as nivolumab, ipilimumab, or dabrafenib plus trametinib, all of which have demonstrated efficacy in the adjuvant setting after the resection of stage III melanoma in large randomized controlled trials^8-10; a trial of adjuvant nivolumab combined with ipilimumab (CheckMate 915) is ongoing. The decision of whether to administer adjuvant therapy should also take into account the timing of disease. Patients with rapid recurrences, even if they have a low volume of disease (one lesion), should be considered for adjuvant therapy, whereas those with a longer disease-free interval and low-volume resected disease may opt to try a period of observation. However, patients with unresectable IT disease or with continued rapid recurrences can present a challenge. As such, a wide spectrum of therapies for patients with isolated IT disease has been developed, including regional therapies, intralesional therapy, systemic therapies, and combinations of these three approaches (Table 1).

Table 1.

Therapies for Patients With In-Transit Melanoma

REGIONAL THERAPY

For patients with IT disease that is confined to an extremity, regional chemotherapy had been the mainstay of treatment since the 1950s. Although 65% to 85% of patients will experience disease recurrence within 3 years after treatment,^11,12 regional chemotherapy continues to provide excellent local control that is not usually matched by systemic treatments. However, regional chemotherapy has not demonstrated a survival advantage compared with systemic treatments. The fundamental principle behind regional chemotherapy involves vascular access and the isolation of the affected limb to administer chemotherapy at doses that are up to 20 times higher than that which can be administered systemically.

Hyperthermic isolated limb perfusion (HILP) involves the direct dissection and cannulation of the major vessels of the extremity, followed by isolation of the limb using a tourniquet and the subsequent heating of the limb to 39°C to 41°C, followed by the infusion of melphalan. Single-center studies have reported an overall response rate (ORR) of 75% to 95%, with complete response (CR) rates of 40% to 70%.^12-15 Although the addition of TNF-α to HILP improved response rates in Europe, with some studies reporting CRs of 70%,^14,16 those findings were not reproduced in the United States.¹⁷ As such, isolated limb perfusion with TNF-α and melphalan is currently only used outside of the United States.

Although isolated limb infusion (ILI) still involves the administration of high-dose chemotherapy to the limb that has been isolated by tourniquet, several features separate it from HILP.¹⁸ Unlike HILP, vascular access in ILI does not involve the complete dissection of the vessels, but uses percutaneous placement of high-flow catheters into the artery and vein of the uninvolved limb that are then advanced into the artery and vein of the involved limb. Although this access is less invasive, it does limit the ability of ILI to treat proximal disease of the thigh, which HILP is able to do by virtue of cannulating the eternal iliac vessels. Furthermore, unlike HILP, which uses a bypass circuit to oxygenate the limb allowing for longer perfusion times and longer OR time, ILI is performed under progressively hypoxic conditions. After placement of the tourniquet, melphalan plus actinomycin D is infused at low flow rates under normothermic to slightly hyperthermic temperatures of 37°C to 40°C. Initial studies of ILI demonstrated an ORR of 84%, with a CR of 38% and median response duration of 13 months overall.¹¹ Subsequent studies yielded ORRs between 45% and 75% with a CR of approximately 30%.^15,19-21 Although both therapies have similar rates of regional toxicity, HILP has higher rates of severe toxicity, including limb-threatening toxicities.²² Lower rates of severe toxicity and shorter operative time make ILI more suitable for patients who are at high risk or who have multiple comorbidities.

Although there have been no direct prospective comparisons of ILI and HILP, it does seem that HILP has superior CR rates and longer durations of response; however, ILI is less invasive and can be performed multiple times. A 2012 study by Chai and colleagues²³ evaluated the use of repeat HILP versus ILI across a multi-institutional setting. In their series, only three (7%) of 44 patients had repeat HILP, whereas 10 (23%) of 44 patients underwent HILP after initial ILI, and 12 (27%%) of 44 underwent ILI after initial HILP. The authors found no significant difference in time to progression after the initial procedure between HILP and ILI, and no difference in survival. On the basis of these findings, Chai and colleagues proposed an algorithm for the use of HILP versus ILI in IT disease, using ILI as initial treatment—except in cases of high-volume or bulky disease—and reserving HILP as a salvage technique for patients with disease progression or rapid recurrence.

INTRALESIONAL THERAPY

Intralesional therapy for IT melanoma has long been explored given the ease of access to lesions and lower risk of systemic side effects. Use of local therapies is especially relevant to patients who are frail or who have multiple comorbidities. Intralesional interleukin-2 (IL-2) has demonstrated promising results but has failed to show an improvement in overall survival.²⁴ BCG (Bacillus calmette-Guerin) is an attenuated live bovine tuberculosis bacillus that has been used as intralesional therapy for the treatment of many malignancies, including melanoma.²⁵ Local injection of PV-10 (Rose Bengal) has also demonstrated prolonged delays in disease progression with an excellent safety profile.²⁶ In phase II trials, intralesional L19–IL-2, an immunocytokine that combines IL-2 and the human monoclonal antibody fragment L19, has been shown to prolong the time to distant metastasis and provide effective regional control of disease progression.²⁷ Injection of Toll-like receptor agonists is also being explored and has been shown to substantially increase leukocyte infiltration and interferon-regulated gene expression.^28,29 In general, a small percentage of patients do experience a response and derive benefit from these therapies, but a survival benefit is unlikely for all patients. Injectable agents as monotherapy do not produce responses in the majority of patients, but intralesional therapy may recruit immune effector cells to the tumor that can be activated by systemic checkpoint inhibitors.^26,29,30

Another strategy for intralesional treatment of IT disease is oncolytic viral therapy. Talimogene laherparepvec (T-VEC), a genetically modified herpes simplex virus type 1, is the first oncolytic virus to be approved for use in patients with advanced melanoma in the United States. Approved in 2015, the virus has been genetically modified to have attenuated viral pathogenicity and to produce granulocyte-macrophage colony-stimulating factor (GM-CSF) to recruit and activate immune cells.³¹ In a phase III trial that examined 436 patients with unresectable, injectable stage IIIB to IV melanoma, treatment with T-VEC was well tolerated overall and resulted in a significantly better durable response rate (16.3% v 2.1%) and ORR (26.4% v 5.7%) compared with intralesional GM-CSF alone.³¹ However, at this time, T-VEC has not been demonstrated to improve survival when used as single therapy.³¹

Other promising viruses for oncolytic viral therapy include ONCOS-102 and PVSRIPO. ONCOS-102 is an adenovirus that is currently being evaluated in clinical trials in combination with pembrolizumab for the treatment of unresectable melanoma (ClinicalTrials.gov identifier: NCT03003676). PVSRIPO is a live attenuated serotype 1 poliovirus vaccine (Sabin) that was modified by the exchange of the entire cognate internal ribosomal entry site with the corresponding segment from human rhinovirus type 2, thus eliminating its neurovirulence through the profound restriction of human rhinovirus type 2 internal ribosomal entry site translation in neurons.³² In preclinical melanoma models, PVSRIPO induced a sustained type I interferon response, activated specific cytotoxic T-cell response, and delayed tumor growth.³² PVSRIPO has demonstrated promising results in the treatment of glioblastoma multiforme, and its use in melanoma trials is planned.³³ Although promising, additional studies and ongoing follow-up is needed to further evaluate the roles of T-VEC and other oncolytic viruses in combination with checkpoint inhibitors.

SYSTEMIC OPTIONS

Systemic therapy should be considered for patients with IT disease and concurrent distant disease. Systemic therapy can also be considered for patients with IT disease plus regional nodal involvement or even for patients with unresectable or high-volume IT disease alone. Checkpoint inhibitor therapies, such as anti–cytotoxic T-cell lymphocyte-4 (anti–CTLA-4) blockade and anti–programmed death 1 (anti–PD-1) blockade, focus on modulating the regulators of T-cell function, thereby removing T-cell inhibitory signals. In KEYNOTE-006, anti–PD-1 therapy using pembrolizumab demonstrated double the rate of progression-free survival (PFS) compared with anti–CTLA-4 therapy with ipilimumab (31% v 14%) as well as a much lower toxicity profile.³⁴ More recently, use of combination anti–CTLA-4 and anti–PD-1 showed significantly longer PFS compared with ipilimumab alone. The CheckMate 067 trial, which compared combination checkpoint inhibitor therapy with a single agent in patients with advanced melanoma, demonstrated an overall survival rate of 58% at 3 years with combination nivolumab plus ipilimumab, 52% with nivolumab alone, and 34% with ipilimumab alone³⁵; however, toxicity from the combination limited widespread use. Anti–PD-1 therapy is currently the backbone of treatment strategies for many patients with IT disease as either monotherapy or in combination with other systemic or intralesional therapies.

BRAF inhibitors (vemurafenib and dabrafenib) plus mitogen-activated protein kinase kinase (MEK) inhibitors (trametinib and cobimetinib)^36-38 are another systemic therapy option for the 50% of melanomas with BRAF mutations.³⁹ The combination of BRAF/MEK inhibition results in ORRs of 67%, although the duration of response and the rapid development of resistance remain problematic for targeted therapy.⁴⁰ Use of BRAF/MEK inhibition may also be useful for the conversion of unresectable IT disease to resectable disease, which is currently being explored in clinical trials of neoadjuvant therapy for patients with stage III disease (ClinicalTrials.gov identifier: NCT02231775). Certainly, patients with IT disease should undergo BRAF mutation testing so that informed decisions can be made about treatment plans.

Additional systemic immunotherapies are also in development. Cell-based immunotherapy through the expansion of tumor-infiltrating lymphocytes, or, more recently, adoptive cell transfer of the patient’s peripheral T cells that have been genetically modified to target cancer-specific antigens via chimeric antigen receptors, have shown promise in the treatment of metastatic melanoma but are not without serious systemic toxicties.⁴¹ Although novel therapies are being explored, long-term follow-up data on the efficacy and toxicity for treatments that are specifically developed for patients with IT disease are not readily available.

COMBINATION THERAPIES

Recognizing that many patients with IT disease have not only local but distant disease at recurrence, current research has focused on the combination of regional chemotherapy with systemic treatments. Mechanistically, regional treatment, by activating the innate immune system via direct tumor cytotoxicity, may augment systemic immunotherapy strategies. Although promising in concept, early investigation into the combination of regional therapy with a nonspecific BRAF inhibitor (sorafenib) yielded disappointing results, with no improvement in response rates and an increase in regional toxicity.⁴² However the combination of systemic immunotherapy with regional chemotherapy does seem to hold promise.⁴³ Preliminary results of a trial of ILI followed by ipilimumab demonstrated a 65% CR rate, 57% PFS at 1 year, and no increase in adverse events as a result of the combination therapy.^44,45

Preclinical studies have also shown that oncolytic viruses can activate the immune system, increasing the susceptibility of both local and distant tumors to subsequent immune checkpoint inhibitor therapy.⁴⁶ A randomized phase II study evaluated the combination of T-VEC with ipilimumab versus ipilimumab alone in 198 patients with unresectable melanoma stage IIIB to IV. The authors observed a significant increase in ORR in the combined therapy arm (39% v 18%), as well as a decrease in size of visceral lesions in both arms (52% v 23%).⁴⁷ Given the toxicities of ipilimumab, T-VEC in combination with anti–PD-1 therapy is currently being investigated. In a phase Ib study, combination T-VEC and anti–PD-1 therapy using pembrolizumab demonstrated a 62% objective response rate and a 33% CR rate. A > 50% reduction was noted in 82% of injected lesions, 43% of noninjected nonvisceral lesions, and 33% of noninjected visceral lesions, which suggests a more systemic antitumor response. Patients who experienced a response to therapy had increased CD8-positive T cells, elevated programmed death-ligand 1 protein expression, and increased gene expression of interferon gamma.⁴⁸ Regional treatments or intralesional therapy in combination with checkpoint inhibition are indicative of the future direction of care for patients with IT disease.

CONCLUSION

Treatment of IT disease depends on several factors: concurrent distant disease, concurrent nodal disease, location of IT disease (extremity or not), volume of IT disease and disease burden, timing of recurrence relative to primary or prior IT melanoma excisions, and overall patient performance status. Evaluation of a patient who presents with IT disease should begin with an assessment of disease burden by physical examination and whole-body imaging.

In the case of either clinically evident nodal or distant disease found concurrent with IT disease, systemic therapy should be strongly considered in the form of immune therapy, targeted kinase therapy, cell-based therapies, or local therapies in combination with a systemic therapy (Fig 1). If there is no other disease observed upon imaging and the lesion is resectable, resection is warranted. Defining resectability can be complex. A general guideline is no more than three lesions, no one lesion greater than 5 cm in size, and lesions that are not rapid, successive recurrences; however, resection can also be considered for isolated symptomatic lesions. If the patient had no prior SLNB or had a prior negative SLNB, IT resection plus SLNB of the most proximal IT lesion may be considered to potentially change the stage, prognosis, and therapy options in a minimally invasive manner. In a medically fit patient, adjuvant therapy with anti–PD-1 or BRAF/MEK combination, depending on mutational status, should be administered after resection. Neoadjuvant regimens, including targeted therapies and immune checkpoint inhibitors (ClinicalTrials.gov identifier: NCT03003676; NCT02519322) for patients with resectable disease, are currently being explored in clinical trials and may be useful options, particularly in patients who present with resectable IT disease plus clinical nodal disease or resectable oligometastatic disease.

Many options are available for the clinical scenario of unresectable IT disease and no clinically evident distant or nodal disease. For lesions on an extremity, ILI can induce CR in up to 30% of patients with minimal toxicity and known durability. Because any intralesional therapy generally requires injections of multiple lesions on multiple days, a patient with an extremity disease burden of more than approximately 25 lesions would likely be a better candidate for ILI as a first approach. T-VEC is approved as first-line therapy for patients with stage III disease, especially for patients who are not candidates for regional chemotherapy, such as those with trunk or head and neck disease. Patients who present with IT disease that has not responded to US Food and Drug Administration–approved and National Comprehensive Cancer Network–recommended therapies are challenging. An algorithm for patients who have experienced failure with standard and first-line treatments is shown in Figure 2. If the patient presents after experiencing multiple treatment failures, repeat ILI, HILP, systemic therapy, or clinical trials of novel intralesional therapies with or without systemic therapy can be performed.

The future of therapy for IT disease will be logical combination therapies that utilize intralesional therapies with systemic therapies. It is unlikely that any intralesional therapy alone will provide sustained tumor control locally and distantly. Although approximately 15% to 25% of patients will benefit from the total control of IT disease because they will never develop distant disease, most patients will develop nodal and distant disease, which decreases their melanoma-specific survival. Control of IT disease is important for patients, especially when lesions bleed, ulcerate, or require ongoing wound care; therefore, therapies that target both local and systemic disease are necessary, and these patients should be discussed at a multidisciplinary tumor board to allow for optimal workup and treatment.

The IT disease model is unique in that the tumor is directly accessible for the injection of therapy, minimally invasive tumor biopsy, and direct visual assessment of tumor response. Individual lesions within one patient have been found to be genetically similar such that one lesion can be representative of the others, which is important for therapeutic decisions.⁴⁹ IT melanoma also represents a unique disease model in which lessons about the tumor microenvironment can be readily learned from appropriately designed studies. There is increasing evidence to suggest that examining the tumor microenvironment via tumor biopsy is critical to the evaluation of immunotherapy.⁵⁰ Discoveries on the basis of tissue correlates can lead to directly translatable therapeutic benefits for patients not only with IT melanoma, but with other forms of metastatic melanoma and solid tumor malignancies.

AUTHOR CONTRIBUTIONS

Conception and design: Jennifer A. Perone, Douglas S. Tyler, Georgia M. Beasley

Collection and assembly of data: Jennifer A. Perone, Georgia M. Beasley

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Contemporary Approaches to In-Transit Melanoma

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jop/site/ifc/journal-policies.html.

Jennifer A. Perone

No relationship to disclose

Nellie Farrow

No relationship to disclose

Douglas S. Tyler

Consulting or Advisory Role: Castle Biosciences, Array BioPharma

Patents, Royalties, Other Intellectual Property: UpToDate

Georgia M. Beasley

No relationship to disclose