Cancers of the skin

14 Cancers of the skin

HM. Hatcher, TV. Ajithkumar

Cutaneous melanoma

Epidemiology

Melanoma is the most aggressive form of skin cancer which is increasing in incidence. In Europe, its incidence has risen by 3–8% per year since the 1960s. The lifetime risk of melanoma in the UK is 1 in 147 for men and 1 in 117 for women. In Australia the risks are significantly higher with lifetime risks of 1 in 25 for men and 1 in 35 for women. Overall survival rates have improved due to early detection but survival of higher stage disease has improved very little in the past 10 years.

The peak incidence of melanoma occurs in middle age (40s) with less than 10% occurring in childhood, although the rates in those under 20 have increased by over 3% in the past 10 years.

Aetiology

Ultraviolent sun radiation (especially UVB) is the main risk factor for the development of cutaneous melanoma. People with a history of blistering sunburns have a 2.5 times higher risk of melanoma. Other surrogate makers of sun sensitivity such as freckling (RR 2.5), burn without tanning (RR 1.7), red hair (RR 2.4) and blue eyes (RR 1.6) also have an increased risk.

Other risk factors are:

• Genetic – people with a strong family history and multiple atypical moles are at the greatest risk of melanoma. Those with an inherited mutation in the CDKN2A and CDK4 genes have 60–90% lifetime risk of melanoma.

• Naevi – Multiple benign naevi (>100) as well as multiple aypical naevi increases the risk of melanoma (RR 11)

• Immunosuppression – Transplant recipients (RR 3) and patients with AIDS (RR 1.5) have increased risk of melanoma

• Previous melanoma – those with a previous melanoma have a 2–10% risk of a further melanoma which increases further with two previous melanomas.

Pathology

There are four histological variants of cutaneous melanoma:

• Superficial spreading is the most common type. This often arises within a pre-existing naevus and is surrounded by a zone of atypical melanocytes that may extend beyond the visible border of the lesion.

• Nodular melanoma (10–15%) presents as symmetrical, uniform dark blue-black lesions. Amelanotic nodular melanomas are often misdiagnosed.

• Lentigo maligna melanoma (10–15%) usually occurs on the sun-exposed areas of head and neck and hands. Clinically these are large (often >3 cm) flat lesions with areas of dark brown or black discolouration. These lesions arise from the premalignant lesion called Hutchinson’s freckle.

• Acral-lentigenous melanomas, as the name suggests (acral – distal), occur on the palms, soles and subungual regions. These lesions occur with the same frequency in whites and non-whites (2–8% of melanoma in whites and 40–60% melanoma of non-whites). These lesions can be easily misdiagnosed as subungual haematoma.

Other forms of melanoma include ocular, mucosal and vulval (p. 236).

The histology of melanomas is very variable depending on the specific subtype and primary site. Many of the features assessed by histology have prognostic value (Box 14.1). In addition to these the pathologist should report the type of melanoma, greatest thickness, radial or vertical growth phase, excision margins and immunohistochemical stains. Immunohistochemical stains used in melanoma include S100 (the most frequently used but also stains benign melanocytes), HMB-45, Mitf, MART-1 and tyrosinase.

Investigations

In the absence of clinical evidence of metastatic disease in regional lymph nodes at presentation, there is no indication for any staging investigations. In patients with thick primary tumours (>4 mm), CT scan may be useful prior to rule out metastatic disease. Studies show that a routine chest X-ray revealed metastasis in only 0.1% whereas 15% showed false positivity. Similarly CT scan showed metastasis in 1.3% while false positivity was 16%.

Patients with >4 mm thick lesion and evidence of nodal metastasis (including micrometastasis) need staging investigations with full blood count, liver function tests, serum lactate dehydrogenase (LDH), chest X-ray and CT scan of chest, abdomen and pelvis.

Tissue diagnosis

Excision biopsy to include 2–5 mm of clinical margin of normal skin with a cuff of subdermal fat is the standard procedure. This helps to confirm diagnosis and provide guidance for subsequent management based on Breslow thickness. Full thickness incisional biopsy from the thickest part of the lesion is acceptable as a mode of diagnosis in certain anatomical areas (e.g. face, ear, palm/sole) and for large lesions. Shave and punch biopsies are not recommended.

Staging

Breslow thickness, ulceration and mitotic count are most important prognostic factors of localized disease whereas Clark level is an independent prognostic factor for melanoma of <1 mm thickness (Box 14.1). Staging is by TNM AJCC system and is related to prognosis (Table 14.1). M1a includes distant skin, subcutaneous or nodal metastases with a normal LDH. M1b is lung metastases with a normal LDH and M1c is with any other metastases or with a raised LDH level.

Table 14.1 TNM staging & 5-year survival of cutaneous melanoma

Staging	5-year survival (%)
Stage IA	95
T1aN0M0 ≤1 mm thickness without ulceration and mitosis <1/mm²
Stage IB	91
T1bN0M0 ≤1 mm thickness with ulceration or mitosis ≥1/mm²
T2aN0M0 1.01–2 mm thickness without ulceration
Stage IIA	77–79
T2bN0M0 1.01–2 mm thickness with ulceration
T3aN0M0 2.01–4 mm thickness without ulceration
Stage IIB	63–67
T3bN0M0 2.01–4 mm thickness with ulceration
T4aN0M0 >4 mm thickness without ulceration
Stage IIC	45
T4bN0M0 >4 mm thickness with ulceration
Stage IIIA
T1-4aN1aM0 Micrometastasis* in 1 node	70
T1-4aN2aM0 Micrometastasis in 2–3 nodes	63
Stage IIIB
T1-4bN1a/2aM0	50–53
T1-4aN1bM0 Macrometastasis** in 1 node	46–59
T1-4aN2bM0 Macrometastasis in 2–3 nodes	46–59
T1-4aN2cM0 in-transit metastases/satellites without nodes
Stage IIIC	24–29
T1-4bN1b/2b/2cM0
Any T N3M0 metastasis in >3 nodes or matted nodes or in-transit or satellites with metastatic nodes
Stage IV
Any T, any N, M1 distant metastasis	7–19%

*micrometastases after sentinel node biopsy; **macrometastases are clinically detectable pathologically confirmed nodes.

Management

Non-metastatic disease

Primary surgery

Surgical excision with an adequate margin is primary treatment of choice for patients with no distant metastasis. The margin is the clinically measured margin during surgery rather than histopathological margin. The extent of the margin depends on the depth of the melanoma and may need to be adjusted for cosmetic or functional reasons. The recommended margins are: 1 cm for lesions less than 1 mm in depth, 1–2 cm for lesions 1–2 mm in depth, and 2 cm for lesions >2 mm in depth.

Management of regional lymph node

The role of elective lymph node excision for node negative patients has being debated for several years. Although initial retrospective studies showed a survival benefit with this approach, four randomized studies failed to show any survival advantage. Debate on the role of elective node dissection has been subsumed by emergence of the sentinel node concept.

The sentinel node (SN) is the node to which the lymph initially drains from a tumour before passing to the other regional nodes. In theory the sentinel node is most likely to contain the tumour cells and if none are present in this node, it is unlikely that other lymph nodes are involved.

The risk of sentinel node metastasis depends on the thickness of lesion. A tumour less than 0.8 mm thick has 1% SN positivity, 0.8 to 1.5 mm has 8%, 1.5–4 mm thickness has 23% and more than 4 mm thickness has 36% risk. Although, sentinel node positivity is proven to have strong correlation with survival (90% 5-year survival for SN negative vs. 56% for SN positive), the role of lymph node dissection for SN positive disease is still evolving. Many agree that there is no proven overall survival benefit from the routine application of SN to patients with cutaneous melanoma. However, there is some suggestion that it may increase the disease free survival.

Patients with >4 mm thick lesions have a predicted incidence of SN positivity of 30–40%. Hence it is reasonable to offer SN in this group of patients to provide prognostic information and selection into clinical trials.

The ongoing MSLT-II trial aims to examine the benefit of complete dissection on survival by randomizing patients with SN positive melanoma to undergo either complete nodal dissection or observation.

Adjuvant treatment

No trial has shown a survival benefit for adjuvant chemotherapy in localized malignant melanoma. Due to the activity of immune treatment in advanced melanoma several trials have examined the role of interferon and vaccines in the adjuvant treatment of melanoma. A pooled analysis of three ECOG studies showed that adjuvant high dose interferon improves relapse-free – survival (about 10% at 5- years) but not overall survival in patients with a high-risk resected melanoma. This group included patients with ≥4 mm thick with no nodal involvement and melanoma involving regional nodes or in-transit metastasis. However, high dose interferon is associated with a number of side effects such as acute constitutional symptoms, chronic fatigue, headache, weight loss, nausea, myelosuppression and depression and most patients will require dose modifications during treatment.

A trial examining the role of a ganglioside vaccine in the adjuvant treatment of high risk melanoma showed no benefit and there were concerns that patients may have done worse than with no treatment.

Management of positive nodes

Patients with metastatic lymph node disease are treated with regional node dissection if feasible and up to 13–59% of these patients do not develop metastatic disease. There is a significant risk of lymphoedema (especially with groin dissections) and patients should wear compression stockings for many months. Some trials are examining the role of adjuvant treatment, such as bevacizumab, after groin dissection. A randomized study has evaluated the role of postoperative radiotherapy (48 Gy in 20 fractions) in patients with a high risk (>25%) of regional recurrence after lymphadenectomy. The two-year lymph node field relapse-free rate was significantly improved with adjuvant radiotherapy compared with observation (82 vs. 65% HR 0.47, 95% CI 0.28–0.67); but without no improvement in 2-year overall survival. The long term toxicity of radiotherapy yet to be reported.

Management of metastatic disease

Metastatic melanoma has a poor prognosis. Most patients with non-visceral metastases survive up to 18 months whereas the median survival of those with visceral involvement or an elevated serum LDH is 4–6 months. Lymph node and skin metastases in the absence of other metastases have the best prognosis (up to 18 months). Those with lung metastases in the absence of other visceral disease have an intermediate prognosis (up to 12 months median survival). Those with other visceral disease have a median survival of less than 6 months which is limited further in the presence of a high and rapidly increasing LDH. Patients with liver metastases from a primary uveal melanoma have a particularly poor prognosis usually limited to less than 3 months.

Treatment of melanoma metastases depends on the site of disease, whether or not it is localized and the overall fitness of the patient. Treatment can be with surgery, radiotherapy, systemic therapies or best supportive care.

Role of surgery

In the presence of an isolated metastasis, especially if the patient has a long disease free interval, e.g. 1 year, surgery may be the best option. The most common sites for surgical resection are skin, brain and lung. Resection of liver metastases is not usually performed as liver metastases are associated with such a poor prognosis. Appropriate staging is important in these patients who undergo surgical resection. PET scans have been used to exclude further distant disease, especially if complex surgery is planned, e.g. resection of cerebral metastases.

Patients with up to three visceral metastases are candidates for surgical resection and the 5-year survival can be more than 20% after complete resection of lung metastases and 28–41% after complete resection of gastrointestinal metastases. Similarly patients with solitary brain metastases can also be candidates for resection. Adjuvant radiotherapy is often used in some of these patients, especially after resection of cerebral metastasis.

Role of radiotherapy

Radiotherapy can also be given after surgical resection of a solitary metastasis and this is particularly the case after resection of a solitary brain metastasis. The treatment options include whole brain radiotherapy (20 Gy in five fractions or 30 Gy in 10 fractions), stereotactic radiosurgery or a combination of both. The optimal radiotherapy in this situation is yet to be defined.

Palliative radiotherapy is useful in bone metastasis and spinal cord compression not amenable to surgical decompression (p. 328). Skin metastases which are rapidly enlarging and about to ulcerate can also be treated with palliative radiotherapy.

Chemotherapy and other systemic treatments in metastatic disease

Dacarbazine is the standard intravenous chemotherapy drug for metastatic melanoma with a response rate of 10–20% and a median duration of response of 3–6 months. It is given 3–4 weekly at doses of 850–1000 mg/m² with nausea as the main side effect. Temozolamide is an oral analogue of dacarbazine and crosses the blood–brain barrier but is more expensive and has no improvement in response rates compared with dacarbazine (Table 14.2). There is no evidence that combination chemotherapy regime is superior to single agent drugs in terms of response rates or survival.

Table 14.2 Response rates (%) for single agents in metastatic melanoma

Only gold members can continue reading. Log In or Register to continue

Related

Stay updated, free articles. Join our Telegram channel

Tags: Specialist Training in Oncology

Jun 18, 2016 | Posted by admin in ONCOLOGY | Comments Off

Full access? Get Clinical Tree

Get Clinical Tree app for offline access

Get Clinical Tree app for offline access

Agent	% Response rate (CR + PR)
Dacarbazine	20
Temozolamide	21
Carmustine (BCNU)	18
Lomustine (CCNU)	13
Cisplatin	23
Carboplatin	16
Vinblastine	13
Paclitaxel