Cutaneous melanoma: Management of local recurrence

INTRODUCTION — Melanoma is an aggressive neoplasm that can metastasize to virtually any organ of the body. For patients with cutaneous melanoma, the prognosis is related to the location, depth, and other biologic properties of the primary tumor, and the presence or absence of locoregional and distant metastatic disease. Surgery represents the principal treatment modality for primary cutaneous melanoma, and adequate excision is important to lessen the risk of a local recurrence. (See "Surgical management of primary cutaneous melanoma or melanoma at other unusual sites" and "Tumor, node, metastasis (TNM) staging system and other prognostic factors in cutaneous melanoma".)

Local recurrence is defined (albeit somewhat arbitrarily) as tumor regrowth within 2 cm of the surgical scar following definitive excision of a primary melanoma with appropriate surgical margins in order to standardize treatment endpoints in clinical trials [1,2]. Recurrences that are more than 2 cm from the primary lesion but are not beyond the regional nodal basin are termed in-transit metastases [3].

The management of locally recurrent melanoma will be discussed here. The management of in-transit metastases is discussed separately. (See "Cutaneous melanoma: In-transit metastases".)

MECHANISMS OF LOCAL RECURRENCE — Local recurrence potentially can arise as a result of one of several mechanisms. These include:

●Incomplete excision of the primary – Local recurrence may be due to incomplete excision of the primary tumor, even in the absence of a "positive surgical margin" in the resection specimen. Sampling error can prevent an accurate analysis of all margins of the surgical specimen. In theory, a wider excision with more generous margins might prevent this form of local recurrence. (See "Surgical management of primary cutaneous melanoma or melanoma at other unusual sites", section on 'Resection margins'.)

●In-transit metastases – Another source of melanoma regrowth in close proximity to the original primary tumor is microsatellite or "in-transit" melanoma nodules that are not excised with the primary tumor mass because they are physically discontinuous from it. The presumed mechanism, lymphatic spread, is supported by the observation that the presence of satellite metastases in a primary excision specimen increases the risk for a subsequent local recurrence [4]. (See "Cutaneous melanoma: In-transit metastases".)

●Second primary melanomas – Misclassification of a new (de novo) primary melanoma that arises at the periphery of a previously excised melanoma may also be confused with a local recurrence. Histologic evidence of a melanoma precursor lesion within a suspected local recurrence should suggest the correct diagnosis. (See "Pathologic characteristics of melanoma".)

●Hematogenous dissemination – A local recurrence can result from hematogenous dissemination with implantation of tumor cells close to the original excision site. A close association between hematogenous dissemination and local recurrence is suggested by data showing a high rate of distant metastases as the next site of relapse in patients with a local melanoma recurrence [5].

RISK FACTORS FOR LOCAL RECURRENCE — The incidence of local recurrence as a first event in disease progression is approximately 3 to 5 percent [5,6]. The majority of local recurrences occur early, but delayed recurrences are not infrequent and are more common with thinner lesions than with thicker ones (table 1).

Several risk factors for local recurrence have been identified:

●Tumor thickness and excision margin – Thick primary melanomas are associated with an increased risk of local recurrence. However, resection of large amounts of normal tissue can affect both function and cosmesis, thus limiting the extent of surgical margins employed in some cases. Recommendations for margin width vary according to tumor thickness and are discussed in detail separately. (See "Surgical management of primary cutaneous melanoma or melanoma at other unusual sites", section on 'Resection margins'.)

●Anatomic site and ulceration – Melanomas arising in the head and neck region and melanomas with ulceration of the primary tumor are at high risk of local recurrence. In the Intergroup Melanoma Trial, the incidence of local recurrence for primary lesions of the proximal extremities, trunk, distal extremities, and head and neck area were 1, 3, 5, and 9 percent, respectively [7]. Among those with melanomas at low-risk sites (ie, proximal extremities or trunk), the risk of local recurrence was significantly increased if ulceration was present (6.6 versus 1.1 percent without ulceration). Those with primary lesions in high-risk locations (head and neck or distal extremities) also had a significantly increased risk of local recurrence if ulceration was present (16.2 versus 2.1 percent).

●Desmoplastic histology – Desmoplastic melanomas have been associated with substantially higher rates of local recurrence than other cutaneous melanomas. For this reason, adjuvant radiation therapy is sometimes indicated following initial resection of these melanomas. Neurotropism, often associated with desmoplastic melanoma, is not an independent risk factor for local recurrence [8]. (See "Radiation therapy in the management of melanoma", section on 'Cutaneous primary lesions'.)

●Lymph node positive disease – Patients who have lymph node involvement (especially multiple positive bulky nodes or extracapsular extension) are at an increased risk of local recurrence as well as in the regional nodal basin. (See "Radiation therapy in the management of melanoma", section on 'Regional lymph nodes'.)

●Satellitosis – The presence of satellite lesions increases the risk of melanoma recurrence [4,5,7,9]. In the Intergroup Melanoma Trial there was a significantly higher rate of local recurrence in patients with satellite lesions (32 versus 2 percent in those without satellite lesions) [5].

LOCAL RECURRENCE AND PROGNOSIS — The development of a local recurrence in a patient with cutaneous melanoma following adequate resection of the primary tumor has historically been associated with a relatively poor prognosis. In the Intergroup Melanoma Trial, the 5 and 10-year survival rates were 9 and 5 percent, respectively, when local recurrence was the first site of failure. In contrast, there was an 86 percent five-year survival rate for those who did not develop a local recurrence [1,5,7]. More recent series that predate the approvals of ipilimumab and BRAF inhibitors suggest 5 and 10-year survival rates in the 50 and 35 percent range, respectively [10]. Given these advances in systemic therapy, the survival of patients with a local recurrence continues to improve.

Whether distant metastases result from local or regional recurrences rather than as a manifestation of aggressive tumor biology is not known. Although it is unlikely that local recurrences are responsible for seeding most distant metastases, local recurrences may seed distant metastases in a subset of patients, and all reasonable measures should be taken to avoid a local recurrence.

Historically, the use of a skin graft to cover the melanoma excision site was advocated to allow early detection of a local recurrence with the goal of improving survival. However, multivariate analysis of the Intergroup Melanoma Trial found that closure by skin graft versus primary closure with full-thickness flap had no impact on survival rates, even after accounting for all other factors that influenced the risk of a local recurrence [7].

One unusual form of "true local recurrence" may be an exception that can be associated with a better prognosis. A local recurrence that is contiguous with the scar and that bears an in situ component that can be excised with a 1 to 2 cm margin is associated with five-year survival rates as high as 89 percent [11]. These lesions have been termed "true local cutaneous recurrent melanoma." Many of these lesions probably represent new primary melanomas occurring within a "field defect" (see 'Mechanisms of local recurrence' above).

MANAGEMENT — Patients who develop a local recurrence of their primary melanoma have a range of therapeutic options available to them. Given the variable availability of these therapies, these patients should be discussed at a multidisciplinary tumor board with expertise in melanoma management whenever possible. These discussions should include both the initial work up as well as the therapeutic options, which can range from simple surgical excision through various regional treatments to systemic therapies potentially used in a neoadjuvant fashion. There are numerous clinical trials available for this group of patients as well. A summary of various staging and therapeutic options are outlined below.

Evaluation

Imaging — Patients with a local recurrence of melanoma should undergo a careful evaluation for other sites of recurrence, including a careful history and physical examination and appropriate imaging studies, such as magnetic resonance imaging (MRI) of the brain and full body positron emission tomography/computed tomography (PET/CT). The evaluation of patients with locoregional recurrence of melanoma is discussed separately. (See "Imaging studies in melanoma", section on 'Approach to patient imaging based on disease site'.)

Lymphatic mapping and SLNB — Appropriate patients with isolated local recurrence or in-transit disease should be evaluated for lymphatic mapping and a sentinel lymph node biopsy (SLNB) because of its prognostic significance and the potential for additional therapeutic options in the setting of a positive SLNB. Factors influencing this decision should be similar to those for patients with an aggressive primary melanoma. (See "Evaluation and management of regional nodes in primary cutaneous melanoma", section on 'SLNB timing and technique' and "Evaluation and management of regional nodes in primary cutaneous melanoma".)

There are two main issues pertaining to SLNB for local recurrence.

●Accuracy after wide excision – Observational studies in patients who underwent SLNB after a wide local excision suggest that an SLNB is accurate and reflects the nodal status of the draining regional basin. Data are as follows:

•In a series of 104 patients from MD Anderson, the sentinel lymph node was identified in 103 (99 percent) and was positive in 18 percent [12]. Over a 51-month median follow-up period there were no nodal basin recurrences in any of the patients.

•In another series of 76 patients, the sentinel lymph node was successfully identified in 75 cases (98 percent) [13]. In this study, 11 patients (15 percent) had a positive SLNB. Among those with a negative SLNB, only 3 of 64 (4 percent) subsequently developed regional lymph node recurrence in conjunction with a systemic recurrence, and one patient had an isolated regional recurrence.

●Prognostic significance – Limited data suggest that SLNB has prognostic significance in patients with local and in-transit metastases. In a study involving 30 patients with local or in-transit recurrence, the median disease-free survival was 16 months when SLNB was positive and 36 months when SLNB was negative [14].

Regarding the role of SLNB in patients with in-transit disease, one study looking at SLNB in 33 patients with in-transit disease found that the technique could be performed successfully in patients who had undergone a previous SLNB, although a previous biopsy can make a repeat procedure more difficult [15]. Using a technique in which the most proximal in-transit lesion was injected with radioactive sulfa colloid, the sentinel lymph node was identified 97 percent of the time and was positive 33 percent of the time. There was a statistically significant difference in the time to development of distant metastatic disease in patients who had a positive sentinel lymph node as compared with those who did not. The authors suggested considering SLNB when patients had a small volume of in-transit disease (<3 lesions) and where surgery was being considered as the primary treatment modality. The role of sentinel lymph node biopsy in patients with more advanced in-transit disease is less clear.

Surgery — Resection is the treatment of choice for most patients with local recurrences of cutaneous melanoma. For patients with a local recurrence and evidence of regional nodal disease either clinically or on imaging studies, discussion at a multidisciplinary tumor board should take place for evaluation of neoadjuvant systemic therapy prior to surgery to remove the local recurrence and further therapeutic nodal dissection.

For patients who are examined at regular intervals following the treatment of a primary melanoma, local recurrences are more likely to be detected at a stage in which they can be successfully managed by surgical resection [16]. Most melanoma surgeons recommend the use of resection margins for recurrences that are sufficient to ensure a grossly and microscopically clear margin, although there are no data that definitively address this issue. Radical surgery employing extensive margins for a local recurrence is not justified. However, even when more conservative margins are used to excise a local recurrence, the surgical defect often requires use of a skin graft because of the previous wide local excision. If surgical excision of the local recurrence cannot be closed primarily and is going to require some type of reconstruction, then patients should be evaluated for alternative initial treatment approaches, such as neoadjuvant treatment or regional therapy. (See "Adjuvant and neoadjuvant therapy for cutaneous melanoma".)

If microscopic lymph node disease is identified on SLNB, controversy exists as to the benefit of completion nodal dissection given the results of the MSLT-II and DeCOG randomized phase III trials. This group of patients should be discussed at a multidisciplinary tumor board to determine the benefit of additional surgery and/or adjuvant systemic therapy. (See "Evaluation and management of regional nodes in primary cutaneous melanoma", section on 'Subsequent management'.)

Adjuvant therapy — Patients with melanoma local recurrences are at high risk for the development of disseminated disease.

Randomized clinical trials are not available to address the value of adjuvant immunotherapy in patients who are rendered disease free following resection of a local recurrence. However, patients who have undergone resection of a local recurrence and have no other evidence of disease could be considered equivalent to stage III and have been included in adjuvant trials; as such, adjuvant therapy (immunotherapy with an anti-programmed cell death 1 [PD-1] agent or targeted therapy for those with a BRAF V600 mutation) should be discussed with patients. (See "Cutaneous melanoma: In-transit metastases", section on 'Adjuvant systemic therapy' and "Adjuvant and neoadjuvant therapy for cutaneous melanoma", section on 'Factors that influence therapy'.)

Isolated limb perfusion — Resection is not an option for some patients because of the extensive nature of their local recurrences. For such patients, hyperthermic isolated limb perfusion (ILP) can be an effective treatment option if the recurrence is located on an extremity. However, increasingly, neoadjuvant systemic approaches are being attempted rather than regional chemotherapy approaches. Regional chemotherapy in the form of ILP or isolated limb infusion tends to be reserved for individuals with large BRAF wild-type lesions who are unable to tolerate systemic immunotherapy. (See "Adjuvant and neoadjuvant therapy for cutaneous melanoma".)

ILP is more commonly applied to patients with extensive in-transit metastases, and these patients predominate in most series. The technique and results with this approach in patients with advanced locoregional disease are discussed separately. (See "Cutaneous melanoma: In-transit metastases", section on 'Regional chemotherapy'.)

Adjuvant ILP — The rationale for applying isolated limb perfusion (ILP) in the adjuvant (ie, postoperative setting) is to eliminate any microscopic residual cells that may remain locoregionally. Although adjuvant limb perfusion decreases the incidence of both in-transit and local recurrences, clinical trials have not demonstrated an improvement in survival. The results with adjuvant ILP are discussed separately. (See "Cutaneous melanoma: In-transit metastases".)

Isolated limb infusion — Because of the technical complexities associated with hyperthermic ILP, investigators have examined intraarterial infusion of chemotherapy alone as an alternative. This technique and results in patients with in-transit metastases and local recurrences are discussed separately. (See "Cutaneous melanoma: In-transit metastases", section on 'Regional chemotherapy'.)

Systemic therapy — Systemic treatment of melanoma results in a complete response in a fraction of patients with disseminated disease. Such an approach is generally reserved for situations in which the local recurrence is not readily resectable. In some cases, initial treatment with systemic therapy (eg, neoadjuvant therapy) can decrease tumor burden and reduce surgical morbidity [17]. (See "Adjuvant and neoadjuvant therapy for cutaneous melanoma".)

Systemic treatment options that have been shown to alter the natural history of metastatic melanoma include therapy with immune checkpoint inhibitors and molecularly targeted therapy against BRAF and MEK in patients whose tumors contain a V600 mutation in BRAF. (See "Overview of the management of advanced cutaneous melanoma".)

Regional therapy with isolated limb infusion or hyperthermic ILP has historically been used in cases of unresectable local recurrences with or without in-transit metastases for limb salvage and has resulted in long-term survival in a fraction of patients. However, systemic therapy has gained preference to regional therapy options. There are limited head-to-head randomized trials comparing regional therapy with systemic therapy approaches. However, there is some evidence that isolated limb infusion followed by immunotherapy may improve antitumor responses [18,19].

Intralesional therapy of cutaneous disease — Intralesional therapy has several goals including local disease control with limited systemic toxicity and the theoretical potential to induce a systemic antitumor immune reaction.

Initially, intralesional Bacille Calmette-Guérin (BCG) was used in an attempt to eradicate dermal and subcutaneous melanoma lesions with the hope of generating a systemic immune response, but severe toxicities led to a marked reduction in the clinical use of BCG. In early reports tumor regressions were seen in approximately 90 percent of injected lesions with regression of uninjected subcutaneous lesions occurring approximately 20 percent of the time. Regression of visceral disease, while reported, was a rare observation using BCG. As the use of BCG became more widespread, some significant toxicities were reported including severe skin reactions, systemic infections, and even anaphylaxis [20].

Compounds that are of interest are talimogene laherparepvec (T-VEC), HLA-B7 gene therapy (allovectin 7) [21], and PV-10 because of the perception that they may generate a specific antitumor immune response and be associated with a higher systemic response rate than is seen with other compounds injected into tumors.

Talimogene laherparepvec — T-VEC is an intralesional therapy that uses an attenuated oncolytic herpes simplex virus containing the granulocyte macrophage colony stimulating factor (GM-CSF) gene. GM-CSF has been used to enhance cellular immunity [22,23]. When T-VEC is injected directly into tumor, its mechanism of action is due to a combination of a direct oncolytic effect from the viral infection and lytic replication, as well as the induction of a systemic immune response.

T-VEC was approved by the US Food and Drug Administration (FDA) to treat unresectable, injectable cutaneous, subcutaneous, and nodal melanoma with limited visceral disease. This approval was based on data from a phase III trial (OPTiM) of 436 patients with unresectable, injectable stage III or IV melanoma with a limited visceral disease burden [24-26]. In this study, T-VEC improved objective and durable response rates, but had similar overall survival compared with GM-CSF. Further details of this study are discussed separately. (See "Cutaneous melanoma: In-transit metastases", section on 'Intralesional therapy (T-VEC)'.)

Additional studies with T-VEC in conjunction with immunotherapy are discussed separately. (See "Systemic treatment of metastatic melanoma lacking a BRAF mutation", section on 'Investigational agents'.)

PV-10 — PV-10 is a sterile, non-pyrogenic solution of 10 percent Rose Bengal disodium which is a fluorescein derivative. Intralesional injection of PV-10 leads to acute necrosis of treated tumor without denaturing the tumor antigens. The local inflammation appears to facilitate exposure of antigenic tumor fragments to antigen presenting cells, leading to the generation of tumor specific effects. In an 80 patient phase II study, up to 20 lesions were injected in each patient [27]. For the entire study population, an objective response (complete or partial) was observed in 41 cases (51 percent).

Other agents — Other intralesional agents that have been studied include the cytokines IL-2, interferon gamma, and interferon alfa. Generally, these agents are associated with high response rates locally at the site of injection but no systemic response.

Radiation therapy — Radiation can provide long-term local control and meaningful palliation in some patients with local recurrences or in-transit disease that cannot be managed surgically. (See "Radiation therapy in the management of melanoma", section on 'Cutaneous primary lesions'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Melanoma screening, prevention, diagnosis, and management".)

SUMMARY AND RECOMMENDATIONS

●Local recurrences of melanoma following adequate and appropriate initial therapy are often a harbinger of disseminated disease. However, aggressive therapy is warranted to minimize morbidity and potentially to increase survival given the favorable outcomes in a subset of patients. (See 'Local recurrence and prognosis' above.)

●Patients presenting with a local recurrence should be carefully evaluated for evidence of distant metastases with imaging such as magnetic resonance imaging (MRI) of the brain and full-body positron emission tomography/computed tomography (PET/CT). (See 'Evaluation' above.)

●For patients with a local recurrence and no other evidence of regional or distant disease, we recommend surgical resection with lymphatic mapping and sentinel lymph node biopsy (SLNB) (Grade 1C). (See 'Lymphatic mapping and SLNB' above and 'Surgery' above.)

•If microscopic lymph node disease is identified on SLNB, controversy exists as to the benefit of completion nodal dissection given the results of the MSLT-II and DeCOG randomized phase III trials. This group of patients should be discussed at a multidisciplinary tumor board to determine the benefit of additional surgery and/or adjuvant systemic therapy. (See "Evaluation and management of regional nodes in primary cutaneous melanoma", section on 'Subsequent management'.)

●For patients with a local recurrence and evidence of regional nodal disease either clinically or on imaging studies, discussion at a multidisciplinary tumor board should take place for evaluation of neoadjuvant systemic therapy prior to surgery to remove the local recurrence and further therapeutic nodal dissection. (See 'Surgery' above and 'Adjuvant therapy' above.)

•For patients who have undergone resection of a local recurrence and have no other evidence of disease, discussion of adjuvant therapy is indicated. (See "Adjuvant and neoadjuvant therapy for cutaneous melanoma".)

●For patients with an unresectable local recurrence (with or without regional nodal recurrence) and no evidence of systemic disease, options include neoadjuvant systemic therapy to downstage the recurrence, isolated limb infusion, isolated limb perfusion, and/or intralesional therapy with talimogene laherparepvec (T-VEC). (See 'Systemic therapy' above and 'Isolated limb infusion' above and 'Intralesional therapy of cutaneous disease' above.)

●For patients with a local recurrence and evidence of systemic disease elsewhere, we offer initial systemic therapy using checkpoint inhibitor immunotherapy or targeted agents. (See "Systemic treatment of metastatic melanoma lacking a BRAF mutation" and "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations".)

•Patients who are not candidates for systemic therapy may be evaluated for the use of intralesional therapy with T-VEC. (See 'Talimogene laherparepvec' above.)

•Whether for curative or palliative intent, the decision to perform surgery in patients with metastatic melanoma is complex. (See "Metastatic melanoma: Surgical management".)