The Washington Manual of Oncology, 3 Ed.

Malignant Melanoma and Nonmelanoma Skin Cancer

Lauren S. Levine • David Y. Chen • Lynn A. Cornelius • Gerald P. Linette

 I. MALIGNANT MELANOMA

  1. Background
  2. Epidemiology. The Surveillance, Epidemiology, and End Results Program (SEER) data demonstrate a steady rise in the incidence of cutaneous melanoma since 1975 and a continued average of 1.8% year-over-year increase between 2002 and 2011. The American Cancer Society estimates that in 2014, approximately 76,100 cases of melanoma will be diagnosed, and 9,710 individuals will die of melanoma. The lifetime risk of being diagnosed with melanoma in the United States is approximately 1 in 50 for whites, 1 in 1,000 for blacks, and 1 in 200 for Hispanics. Overall, more men account for new cases of melanoma than women (27.7 vs. 16.7 new cases per 100,000).
  3. Risk factors. There are clear genetic and environmental determinants of melanoma risk. One study found that mutations in the CDKN2A tumor suppressor gene are present in 39% of families with multiple occurrences of melanoma, while another population-based study found that though CDKN2A mutations were associated with greater risk of first degree relatives with melanoma, only one out of 18 families with three or more affected first-degree relatives carried CDKN2A mutations, suggesting other heritable factors. Low penetrance gene variants associated with increased melanoma risk include melanocortin 1 receptor (MC1R), tyrosinase (TYR), cyclin-dependent kinase 4 (CDK4), micropthalmia transcription factor (MITF), BAP1, and others. New genetic risk factors are being described with the proliferation of next-generation sequencing, including a recent study that identified a nongenic polymorphisms affecting the regulatory region in telomerase (TERT) in a family with multiple family members with melanoma. Despite identification of genetic risk factors, the utility of their detection is not established and genetic testing is not a routine practice in the clinical setting.

 The most significant environmental exposure that drives melanoma, and nonmelanoma skin cancer, risk is ultraviolet (UV) exposure. The WHO has classified UV radiation between 100 and 400 nm as a known carcinogen. UV exposure interacts with genetic risk factors, including fair skin, red hair (MC1R variants), and UV sensitivity syndromes like xeroderma pigmentosum. Fair-skin people are at higher risk of melanoma from UV exposure. Additionally, SEER registry data demonstrate that prior history of melanoma confers 10-fold risk of subsequent melanoma compared to the general population, likely reflecting a confluence of genetic and environmental causes. Other risk factors include increased number of nevi (>50), history of greater than five clinically atypical nevi, large congenital nevi, and history of immune suppression with solid organ transplantation.

  1. Primary cutaneous melanoma
  2. Diagnosis. Cutaneous melanomas commonly arise in the absence of a clinically apparent precursor, although in some instances benign nevi are associated with melanoma on histologic examination. Patients may report the appearance of a new skin lesion or change in an existing lesion and will occasionally note associated symptoms such as itching and bleeding. Nonpigmented, or amelanotic, primary lesions comprise approximately 5% of cutaneous melanomas.
  3. Physical examination. While evaluating a pigmented skin lesion, the ABCD morphologic criteria are helpful, but not absolute.
  • Asymmetry. One half of the lesion does not match the other.
  • Border irregularity. The lesion has ragged or notched edges.
  • Color variegation. Pigmentation is a heterogeneous mixture of tan, brown, or black. Red, white, or blue discolorations are particularly of concern.
  • Diameter. Larger than 6 mm in diameter.
  • Evolution. Any change in clinical characteristics of a lesion noted by the patient or physician.

Particular attention should be given to lesions that are evolving by clinical documentation (i.e., written or photographic records) or by patient report. Together, these sets of criteria are sometimes known as the ABCDE’s of melanoma. Lesions with one or more of these attributes should be brought to the attention of a physician, preferably a dermatologist, and evaluated for the possibility of melanoma. Other characteristics such as itching, bleeding, and the presence of ulceration should also prompt a careful evaluation for melanoma. In addition to examination of the lesion in question, a comprehensive skin examination by a dermatologist is critical in evaluating and monitoring patients with multiple or atypical nevi, a history of excessive sun exposure, or a history of melanoma or nonmelanoma skin cancer. Full body examination is essential, including scalp, hands and feet, genitalia, and oral cavity.

  1. Biopsy. The differential diagnosis of a pigmented skin lesion includes an atypical nevus, a benign growth such as melanocytic nevus, solar lentigo, seborrheic keratosis, angioma, and other malignant growths such as basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). When melanoma or other malignant lesion is a consideration, a biopsy without delay is required to establish a diagnosis.
  • Excisional biopsy. Full thickness removal of the entire clinical lesion with 1 to 3 mm margins is optimal for diagnosis and accurate staging by Breslow’s depth. Avoiding wider margins facilitates accurate sentinel lymph node mapping if later required.
  • Incisional biopsy. For large lesions or lesions on special sites like the palms and soles, face, ears, or digits, full thickness incision or punch biopsy of the thickest clinical portion may be appropriate.
  • Deep shave (Saucerization). Wide sampling is preferred in superficial lesions such as lentigo maligna, where atypical melanocytes may extend beyond the clinically observed lesion. Superficial shave biopsy is not recommended for any lesion suspected to be melanoma.
  1. Histologic reporting and classification. Breslow’s thickness in millimeters, presence or absence of histologic ulceration, dermal mitotic rate per square millimeter, and presence or absence at the lateral or deep margins comprise the bare minimal elements that should be reported with the histologic evaluation of melanoma. Reports may include additional elements encouraged by the American Academy of Dermatology such as the presence or absence of regression, microsatellitosis, tumor infiltrating lymphocytes, lymphovascular invasion, neurotropism, or whether there is vertical growth phase. The pathologist may also report the histologic subtype, which includes superficial spreading melanoma, nodular melanoma, lentigo maligna melanoma, and acral lentiginous melanoma. Superficial spreading melanoma is the most common subtype comprising 75% of all melanomas, while lentigo maligna comprises 10% to 15% and is thought to have an extended radial growth phase. Nodular melanomas are by definition in vertical growth phase. Acral lentiginous melanoma is the least common type and characteristically arises on specialized sites including palmar, plantar, and subungual locations. Aside from the four dominant subtypes, there are rare variants including nevoid melanoma and desmoplastic melanoma. Although histologically distinct, the subtype does not affect staging as it does not influence management or prognosis with the exception of pure desmoplastic melanoma, with which sentinel node biopsy may not be indicated.
  2. Diagnostic dilemmas. Distinguishing melanoma from benign growths can be challenging, despite thorough clinical exam and adequate sampling of the lesion. Immunohistochemistry may be utilized to highlight cells of melanocytic origin, including S100, MART-1, and HMB-45. These antigens are not specific to melanoma but can highlight the architecture of the lesion as well as aid in the identification of nodal metastases. In certain cases, further testing for chromosomal aberrations and copy number variability with fluorescence in situ hybridization (FISH) or comparative genomic hybridization (CGH) can help distinguish benign from malignant lesions.

Staging. The most commonly used staging system is found in the AJCC Cancer Staging Manual, 7th edition. The T stage is defined by the tumor thickness (T1—≤1 mm, T2—1.01 to 2 mm, T3—2.01 to 4 mm and T4—>4 mm). The presence of ulceration defines the b substaging. The N stage is defined by the number of lymph nodes involved (N1—1 lymph node, N2—2 to 3 lymph nodes and N3—4 or more lymph nodes, matted lymph nodes or in-transit metastases). The M stage may be further subdivided into M1 (involvement of skin, subcutaneous tissue, or distant lymph nodes with normal LDH), M2 (lung metastases with normal LDH), and M3 (metastases to other organs or any metastasis with elevated LDH). The melanoma stages are as follows: IA (T1a), IB (T1b or T2a), IIA (T2b or T3a), IIB (T3b or T4a), IIC (T4b), III (N1, N2, or N3), IV (M1). Although many of the staging criteria remained the same, the new staging system reflects the rise in utilization of sentinel lymph node biopsy technique and the attendant increase in detection of micrometastases. Importantly, microstaging now distinguishes mitotic rate less than 1/mm2 or 1/mm2 or greater as a separate staging criterion in determining T1a from T1b lesions, respectively, in addition to the existing criterion regarding the presence or absence of ulceration. The most important prognostic factors in the staging of melanoma are the thickness of the primary lesion measured in millimeters, the presence of histologic ulceration, the mitotic rate, and regional lymph node involvement. The thickness of the primary melanoma is known as the Breslow thickness and is measured in millimeters from the top of the granular layer in the epidermis to the base of the deepest tumor nest in the dermis. The Breslow thickness cutoffs for primary tumor classification are 1.0, 2.0, and 4.0 mm, while mitotic activity and ulceration modify tumor staging. Tumor staging, in combination with clinical findings, guide the next steps of the staging workup.

  1. Sentinel lymph node biopsy. Stage 0, I, and II melanoma is localized to the skin, while stage III melanoma denotes regional metastasis, which is detected by clinical exam or by sentinel lymph node biopsy. Lymphoscintigraphy and sentinel lymph node biopsy is performed at the time of wide-local excision and offers prognostic value to patients with primary melanoma thicker than 1.0 mm. This is supported by multiple studies and recently reaffirmed in the final analysis of the Multicenter Selective Lymphadenectomy Trial-1 (MSLT-1). Generally, sentinel lymph node biopsy is not recommended for primary melanomas less than 0.75 mm, and there is no general consensus for melanomas between 0.76 and 1.0 mm.
  2. Imaging. Routine imaging is not recommended in stage I or II disease unless used to evaluate specific signs and symptoms. The exception is ultrasonography of a nodal basin for an indeterminate lymph node clinical exam, which can help guide decisions for fine-needle aspiration (FNA) or sentinel biopsy. For stage III disease as determined by sentinel biopsy, clinically positive nodes, or in-transit metastases, baseline contrast-enhanced CT exam is recommended, with or without positron emission tomography with computed tomography (PET-CT) or MRI, based on clinical context. For suspected stage IV disease, in addition to CT of the chest, abdomen, and pelvis, gadolinium-enhanced brain MRI is recommended in the initial staging because of its increased sensitivity for detecting small posterior fossa lesions (<1 cm) compared to head CT. PET-CT is also appropriate for initial staging.
  3. FNA. Suspected regional metastatic disease by clinical exam or imaging should be evaluated histologically by FNA. In the appropriate context, this can be done for suspected stage IV disease except when archival tissue is not available for genetic testing. In this instance, biopsy is preferred over FNA.
  4. Lactate dehydrogenase. Elevated serum level of LDH is an independent predictor of poor outcome in stage IV disease and defines the designation of M1c disease. The two-year survival for normal versus elevated LDH at time of staging is 40% versus 18%, respectively, for stage IV (J Clin Oncol 2009;27:6199). For this reason, LDH levels should be evaluated in the initial staging workup of patients with stage IV disease. Monitoring LDH levels in patients with loco-regional disease is not recommended.
  5. Treatment of localized disease
  6. Wide local excision. In stage 0, I, and II disease, wide local excision of the primary lesion with appropriate clinical margins provides the greatest chance of local control. Margins of 0.5 to 1.0 cm are recommended for melanoma in situ. A margin of 1.0 cm is adequate for primary melanomas with a Breslow thickness of 1 mm or less, while melanomas between 1.01 and 2.0 mm thickness require a 1 to 2 cm margin. Any melanoma with a Breslow thickness of greater than 2 mm requires a 2.0 cm clinical margin. More aggressive margins than those recommended have not been demonstrated to improve survival. In the case of stage III disease with clinically evident lymph nodes, the same margins apply for optimal local control, while stage III in-transit disease should be completely excised with clear margins if possible.
  7. Nonsurgical therapy. Although surgical excision is standard of care for in situ melanoma, topical imiquimod may be considered particularly for melanoma in situ or in cases of lentigo maligna when surgical cure is not achievable.
  8. Treatment of advanced melanoma
  9. Adjuvant therapy. There is no proven benefit for any adjuvant therapy given to patients with low-risk (stage IA) or intermediate-risk (stage IB, IIA) melanoma. Interferon α 2b (Intron) was granted U.S. Food and Drug Administration (FDA) approval in 1995 for administration to patients with surgically resected stage IIB, IIC, and III (high-risk) melanoma. The approved schedule is 1 year of adjuvant treatment given intravenously for the initial 4 weeks at 20 MU/m2 each day (Monday to Friday), followed by subcutaneous administration at 10 MU/m2 for 48 weeks given three days per week. We routinely hydrate patients with 500 mL saline before each i.v. dose of interferon during weeks 1 to 4. Patients are premedicated with acetaminophen 650 mg orally. Three randomized controlled clinical trials have been performed using the FDA-approved schedule and the results of Eastern Cooperative Oncology Group (ECOG) 1684, ECOG 1690, and ECOG 1694 have been published. A recent update of the three trials (Clin Cancer Res 2004;10:1670) confirms the durable benefit with improved relapse-free survival (RFS) at a median follow-up of 12.6 months for patients given interferon when compared to the control group.

 Pegylated-interferon α 2b (Sylatron) can be administered as adjuvant therapy to patients with microscopic or macroscopic lymph node involvement with melanoma that has been surgically resected. The greatest risk reductions were observed in patients with ulceration and stage IIb/III-N1. The efficacy of IFN/PEG-IFN is lower in stage III-N2 patients with ulceration and uniformly absent in patients without ulceration. The recommended dose is 6 µg/kg/week subcutaneously for 8 doses, followed by 3 µg/kg/week for up to 5 years (J Clin Oncol 2012;30:3810). There is no role for adjuvant cytotoxic chemotherapy or adjuvant high-dose interleukin 2 (IL-2) for treatment of surgically resected melanoma.

 Ipilimumab as adjuvant therapy for high-risk resected stage III patients has been evaluated in two large randomized controlled clinical trials (ECOG 1609 and EORTC 18071). Final RFS data released at ASCO 2014 for EORTC 18071 demonstrated improved median RFS with adjuvant ipilumumab (17.1 months for placebo to 26.1 for ipilimumab (HR 0.75 [CI 0.64 to 0.90], p = 0.0013)). Moreover, the 3-year RFS rate of 46.5% in the ipilimumab arm is significantly improved compared with 34.8% in the placebo arm. Data on overall survival (OS) are still not available.

 Targeted agents such as serine/threonine-protein kinase B-Raf (BRAF) inhibitors as single agents or in combination with MEK inhibitors are also currently being evaluated as adjuvant therapy in high-risk surgically resected stage III cutaneous BRAF V600E/K-mutated melanoma. The randomized, placebo-controlled phase III studies are ongoing and no data are currently available.

  1. Interferon side effects and toxicities. The side effects and toxicities of interferon are significant and all patients should be counseled before the initiation of therapy (Oncologist 2005;10:739). Virtually all patients experience fatigue and many experience fevers, chills, and diaphoresis (70% patients, grade 3 to 4). Myelosuppression, hepatotoxicity, and neurologic symptoms are frequent. Depression can be severe and precautions should be taken with appropriate referral to mental health professionals. The use of selective serotonin reuptake inhibitors (SSRI) is recommended in suitable patients (N Engl J Med2001;344:961). Approximately 50% of patients have treatment delay or dose reduction during the initial 4 weeks of induction therapy. Selection criteria used currently in our practice include patients aged 60 or younger with no other significant medical illness, who understand the risks and benefits of treatment. Excellent guidelines to assist in the management of toxicities and side effects have been published. Finally, there is no role for administration of interferon concurrent with radiation.
  2. Metastatic disease. Regional and distant lymph nodes, skin, lung, liver, and brain are the most common distant sites of metastases from cutaneous melanoma. Prognosis depends on the sites of metastases with brain and hepatic metastases having the shortest survival followed by lung metastases; nodal and skin metastases have the most favorable prognosis. Patients with regional nodal disease or a single distant site (including brain or lung, particularly with pulmonary disease) should be considered for surgical resection. Complete surgical resection of nodal disease can afford significant long-term survival in many patients. The final analysis of the MSLT-1 data attempted to address whether reflex completion lymphadenectomy in sentinel node positive melanoma patients provided melanoma-specific survival benefit compared to delayed lymphadenectomy. The data for intermediate thickness melanomas suggest survival benefit from reflex lymphadenectomy, though owing to post hoc subgroup analysis these do not provide definitive evidence (N Engl J Med 2014;370:599). Despite this controversy, current guidelines support completion lymphadenectomy for sentinel-node-positive nodal basins (J Clin Oncol 2012;30:2912). The forthcoming MSLT-2 trial will address whether sentinel-node-positive patients benefit from completion lymphadenectomy or from nodal surveillance with ultrasound.

 The treatment of metastatic melanoma has improved substantially since 2011. Advances in immunology as well as molecular oncology provided the foundation for therapeutic strategies that have had a profound effect on patient care. A pivotal discovery in 2002 identified the BRAF V600 mutation as a critical driver mutation that is present in approximmately50% of cutaneous melanoma samples irrespective of geography (Nature2002;417:949). Among patients with BRAF-mutated cutaneous melanoma, the BRAF V600E mutation is detected in 80% to 85% patients and the BRAF V600K detected in approximately10% of cases. It appears that non-V600E genotypes occur more frequently in older (>65 years of age) patients. Rare non-V600E/K (exon 15) mutations are not detected by the FDA-approved companion diagnostic (real-time qPCR) Cobas 4800 BRAF V600 mutation test (Roche/Genentech) and THxID-BRAF mutation test (bioMerieux/GSK). Therefore, in certain cases, DNA sequencing of exon 15 is warranted. An emerging method to detect the BRAF V600E mutation is the implementation of the immunohistochemical (IHC) assay using a monoclonal antibody specific for the V600E-mutated protein. The VE1 monoclonal antibody (specific for BRAF V600E) demonstrates 97% sensitivity and 98% specificity in detecting the V600E mutant protein by IHC assay. It is imperative that BRAF inhibitors are not administered to patients with melanoma that is BRAF wild type (no mutation detected) as there is evidence of a paradoxical activation of the MAPK pathway in various cell lineages, including melanocytes. For selected BRAF wild-type patient groups (e.g., mucosal, acral, and chronic sun-damaged skin), KIT mutation testing is performed.

 Since 2011, ipilumumab, vemurafenib, dabrafenib, and trametinib have received regulatory approval for unresectable or metastatic (stage III/IV) melanoma and each single agent has been shown to prolong survival in randomized clinical trials compared to dacarbazine. In addition, the combination of dabrafenib and trametinib was approved in 2014 based on improved response rate and the median duration of response compared to dabrafenib alone. Pembrolizumab was granted accelerated approval in 2014 based upon the tumor response rate (24%) and the durability of response for the treatment of metastatic melanoma following ipilimumab treatment. Nivolumab subsequently received accelerated FDA approval for the same indication shortly thereafter in late 2014 based upon tumor response rate (40%) and one year survival rate (72.9%). Despite the introduction of these novel agents, clinical trial participation remains the best option for most patients. Prior to 2011, dacarbazine and IL-2 were regarded as the standard of care for patients with newly diagnosed metastatic melanoma. High-dose IL-2 is recommended at some specialized centers in the United States as first line therapy based on the observation that 3% to 5% of selected patients can attain a durable complete remission. Dacarbazine and combination cytotoxic regimens such as carboplatin plus paclitaxel are generally reserved for patients that have failed or are not candidates for checkpoint inhibitors or targeted agents such as BRAF inhibitors.

 Ipilimumab is an immune-modulatory mAb that promotes T cell activation by blocking the interaction of CTLA-4 with its ligands CD80/CD86 (B7 family). Ipilimumab is a human IgG1 kappa antibody specific for human CTLA-4. Since CTLA-4 is a negative regulator of T cell activation, ipilimumab allows T cell activation and proliferation to continue after antigen stimulation by interfering with a homeostatic checkpoint that normally inhibits T cell growth. Ipilimumab was approved in March 2011 for use in patients with unresectable or metastatic melanoma administered every 3 weeks (at 3 mg/kg i.v. over 90 minutes) times four doses with the initial restaging examination done 2 weeks after the fourth or final dose. The overall response rate is 10% to 15%, with a small subgroup of patients exhibiting disease progression at the initial assessment prior to tumor regression. Ipilimumab has been shown in randomized clinical trials to prolong OS in patients with metastatic melanoma (New Engl J Med 2010;363:711). A recent pooled analysis of melanoma patients treated on various clinical trials with ipilimumab confirms a 3-year survival rate of 22%, suggesting a durable long-term benefit in this group. Ipilimumab can cause serious side effects leading to severe, and sometimes fatal, autoimmune reactions resulting in dermatitis, colitis, hepatitis, endocrinopathy, and neuropathy. Additional less common autoimmune toxicities such as nephritis, pneumonitis, meningitis, pericarditis, uveitis, iritis, and hemolytic anemia have been reported. Patients should be evaluated at the start and prior to each ipilimumab dose along with appropriate laboratories (including LFTs and thyroid function tests). Adverse event management guidelines have been issued by the drug manufacturer and should be referred to for specific recommendations and treatment algorithms of specific toxicities. Permanently discontinue ipilumumab and initiate systemic high-dose corticosteroid therapy for severe immune-mediated reactions.

 Pembrolizumab (Keytruda) is a checkpoint inhibitory mAb that blocks the interaction of PD-1 with its ligands (PD-L1/PD-L2). Pembrolizumab is a humanized IgG4 kappa antibody specific for human PD-1. By binding to the PD-1 receptor and blocking the interaction with the receptor ligands, pembrolizumab releases the PD-1 pathway-mediated inhibition of the T cell response, including the anti-tumor immune response. Results from a randomized dose comparison phase 1 study (2mg/kg versus 10 mg/kg) in melanoma patients who progressed after ipilimumab reports ORR 26% (RECIST v1.1) confirmed by independent central review (Lancet 2014; 384:1109). The most common drug-related adverse events of any grade in the cohort were fatigue, pruritus, and rash. The rate of immune-related serious adverse events is 2%. The FDA approved dose is 2mg/kg intravenously over 30 min every 3 weeks.

 Nivolumab (Opdivo) is a checkpoint inhibitory mAb that also blocks the interaction of PD-1 with its ligands (PD-L1/PD-L2). Nivolumab is a fully human IgG4 kappa programmed death 1 (PD-1) immune-checkpoint–inhibitor antibody that selectively blocks the interaction of the PD-1 receptor with its two known programmed death ligands, PD-L1 and PD-L2, disrupting the negative signal that regulates T-cell activation and proliferation. Results from a randomized controlled phase 3 trial in patients with untreated advanced BRAF wild type melanoma reported a 40% ORR with nivolumab compared to 13.9% with dacarbazine (NEJM on line published November 16, 2014). The 1 year survival rate was 73% in the nivolumab group versus 42% in the dacarbazine group (Hazard ratio 0.42, p<0.001). The most common drug-related adverse events of any grade in the nivolumab cohort were fatigue, pruritus, and nausea. The rate of immune-related serious adverse events is ~5%. The FDA approved dose is 3 mg/kg intravenously over 60 min every 2 weeks. Permanent discontinuation of pembrolizumab or nivolumab and initiation of systemic-high dose corticosteroid therapy is advised in the setting of severe immune-mediated adverse events.

 Vemurafenib (Zelboraf) is an oral kinase inhibitor for the treatment of patients with unresectable or metastatic BRAF V600E-mutated melanoma. The recommended dose is 960 mg every 12 hours with or without food. Vemurafenib is available as 240 mg tablets. Extensive clinical testing in BRAF V600-mutated melanoma confirms a 50% to 60% response rate, with a median PFS of 6 to 7 months and median OS 13 to 14 months (Lancet Oncol 2014;15:323). Vemurafenib also has clinical activity in patients with the BRAF V600K mutation. The side-effect profile includes warning for cutaneous toxicities such as hypersensitivity reactions (including pruritus, fever, and erythema), photosensitivity, alopecia, rash (including Stevens–Johnson syndrome), and development of invasive SCC as well as keratoacanthoma, skin papilloma, and new primary melanoma. Other known toxicities include arthralgia, myalgia, nausea, fatigue, uveitis, blurred vision, hepatotoxicity with elevated liver function tests, and QT prolongation. Vemurafenib should not be administered concurrently with ipilimumab due to hepatotoxicity. Dermatology assessment is recommended at baseline and every 2 months while taking vemurafenib due to the various cutaneous side effects. Likewise, monitoring ECGs at baseline and at the recommended intervals is recommended to assess for QT prolongation >500 ms.

 Dabrafenib (Tafinlar) is an oral kinase inhibitor for the treatment of patients with unresectable or metastatic BRAF V600E/K-mutated melanoma. In January 2014, Dabrafenib in combination with trametinib was approved for the treatment of BRAF V600E/K-mutated melanoma based on response rate and the median duration of response (see below). The recommended dose is 150 mg BID. Dabrafenib is supplied as 50- and 75-mg capsules. Dabrafenib shows similar efficacy compared to vemurafenib with 50% to 60% response rate. The side-effect profile is similar; however, there are several important differences. There is a potential risk of hemolytic anemia in patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency. Pyrexia is more common with dabrafenib. Serious febrile reactions and fever of any severity complicated by hypotension, rigors or chills, dehydration, or renal failure can occur and dabrafenib should be held for fever >101.3 F. Similar to vemurafenib, dermatology assessments for skin checks at baseline and every 2 months are recommended. It should be noted that second (noncutaneous) malignancies such as KRAS-mutated pancreatic adenocarcinoma have been diagnosed in patients taking dabrafenib and are thought to be due to the paradoxical activation of the MAP kinase pathway in RAS-mutated tumors, similar to the experience with vemurafenib-treated patients.

 Trametinib (Mekinist) is an oral MEK inhibitor for the treatment of patients with unresectable or metastatic BRAF V600E/K-mutated melanoma. Trametinib as a single agent is not indicated for patients that received prior BRAF inhibitor therapy. At this time, the primary use of trametinib is in combination with dabrafenib for treatment of BRAF V600E/K-mutated melanoma. The recommended dose of trametinib either as a single agent or in combination with dabrafenib is 2 mg daily. Trametinib is supplied as 0.5-, 1-, and 2-mg tablets. Published reports with single-agent trametinib demonstrate a 20% to 25% response rate in patients with BRAF V600E/K-mutated melanoma. Rare toxicities include cardiomyopathy, retinal vein occlusion or retinal detachment, interstitial lung disease, and hyperglycemia. The most common adverse events are rash, diarrhea, and lymphedema.

 The combination of a BRAF inhibitor (such as dabrafenib) and a MEK inhibitor (such as trametinib) is the preferred systemic treatment for patients with unresectable or metastatic BRAF V600E/K-mutated melanoma (NEJM 2014; 371:1877). The recommended dose is dabrafenib 150 mg BID and trametinib 2 mg daily taken 1 hour before or 2 hour after a meal. Treatment is continued until disease progression or unacceptable toxicity. The most common adverse reactions (>20%) for the combination include pyrexia, chills, fatigue, rash, nausea, vomiting, diarrhea, abdominal pain, peripheral edema, cough, headache, arthralgia, night sweats, decreased appetite, constipation, and myalgia. Major hemorrhagic events such as intracranial or gastrointestinal, venous thromboembolism/pulmonary embolism, and cardiomyopathy occur at higher incidence (5% to 10% incidence) compared to patients receiving dabrafenib alone (none reported) in clinical studies. Left Ventricular Ejection Fraction (LVEF) should be evaluated at baseline and at one month and then every 2 to 3 months for patients receiving the combination (or single-agent trametinib). Frequent dermatology evaluation is recommended to assess cutaneous toxicities and new skin lesions.

  1. Investigational therapies. Clinical investigation is active for the treatment of metastatic melanoma in two broad areas: new immunotherapy approaches and drug-resistant BRAF V600-mutated tumors. Monoclonal antibodies directed to PD-1 as well as PD-L1 are being studied in patients with metastatic melanoma and the results are encouraging with 30% to 40% response rates reported in ipilimumab-naïve as well as ipilimumab refractory patients. A preliminary signal of activity has been reported for the combination of ipilimumab and nivoloumab, an antiprogrammed cell death (PD1) monoclonal antibody. The development of new inhibitor combinations targeting the MAPK pathway is a priority. Combination therapies with BRAF inhibitor and MEK inhibitor together with checkpoint inhibitor monoclonal antibodies (such as anti-CTLA-4 and anti-PD-1/PD-L1) show promise for BRAF-mutated melanoma and are currently under study in phase 1 trials. Adoptive antigen-specific T cell therapy is being investigated at major academic centers. Clinical trial participation remains the best option since most patients will ultimately develop disease progression, and thus, more effective treatments options are urgently needed.
  2. Brain metastases. The expected survival of patients with intracranial metastases depends on the number of metastatic lesions as well as the presence of neurological symptoms and performance status. A small percentage (<5%) of patients will develop a Central Nervous System (CNS) only relapse. Patients with greater than 3 intracranial melanoma metastasis and focal neurological deficits have an especially poor prognosis. Despite definitive treatment with corticosteroids and whole-brain radiotherapy in patients with multiple melanoma brain metastases, the median survival is 3 to 4 months. In patients who undergo craniotomy and complete surgical resection of a solitary intracranial metastasis, the median survival is 9 months. Outcomes with stereotactic radiosurgery as definitive treatment for 1 to 2 brain metastases appear equivalent to surgery; however, data from a randomized clinical trial are lacking. Noteworthy is the recent observation that BRAF V600E/K-mutated melanoma in the CNS is sensitive to oral BRAF inhibitors. Both vemurafenib and dabrafenib have been studied in this population when given as single agents. However, caution is advised and the use of BRAF inhibitor with concurrent radiation is not recommended pending formal clinical evaluation. In our experience, it appears safe to administer BRAF inhibitor (with a 10 day window) prior to or after radiotherapy in patients with good performance status; however, additional clinical study in this patient population is warranted.
  3. Follow-up. Patients with a history of melanoma should be followed closely with detailed dermatologic and lymph node examinations. They should be taught skin self-examination, as they are at increased risk for a second primary melanoma, as well as recurrence of disease. In addition, these patients need to be counseled regarding the daily use of a broad-spectrum sunscreen that blocks both UVA and UVB. Patients should also be taught sun-avoidance strategies such as avoiding the mid-day sun and wearing protective clothing. Patients diagnosed with melanoma of any stage are not eligible to donate blood, tissue, or solid organs.

Patients with stage 0 melanoma should be followed with periodic skin examinations for life. Current recommendations for stage IA to stage IIA with no evidence of disease are to have a history and physical examination (H&P) every 6 to 12 months for the first 5 years, then annual skin examinations for life. Routine imaging is not recommended and should be considered only as the clinical scenario dictates. Stage IIB and greater melanoma with no evidence of disease warrants clinical examinations every 3 to 6 months for the first 2 years after diagnosis, then every 3 to 12 months for 3 years, and then annually. Consider radiographic evaluation with chest radiograph, CT, or PET-CT every 4 to 12 months, as well as a brain MRI every 12 months, to assess for metastatic or recurrent disease. Routine radiologic screening in stage IIB and higher is not recommended if rendered no evidence of disease after 5 years, unless symptoms warrant imaging.

  1. Special considerations
  2. Melanoma of unknown primary. Patients may present with metastatic disease without an identifiable primary cutaneous melanoma. Cases of melanoma with unknown primary represent less than 5% of melanomas overall. Most patients present with subcutaneous disease or localized lymph node metastasis clinically manifesting as lymphadenopathy; however, patients with solitary pulmonary metastasis as well as solitary brain metastasis are frequently found to have metastatic melanoma after pathological examination. In these rare instances, BRAF and KIT mutational analysis should be requested. All patients should have a thorough evaluation including examination of the skin, scalp, perineum, eyes, and mucosal membranes, as melanocytes are also present in the eye (conjunctiva and uvea), gut, inner ear, and nasopharynx. Numerous studies have demonstrated that these patients have the same survival as patients with known primaries according to the stage of disease and should be treated accordingly.
  3. Mucosal melanoma. Mucosal melanoma is rare and represents less than 1% of all melanomas. Melanomas can occur on any mucosal surface including the nasopharynx, oral mucosa, larynx, vulva, rectum, and anus. These tumors are generally advanced at the time of presentation and, therefore, prognosis is poor. Treatment is wide local excision with negative histologic margins. We recommend molecular testing for BRAF and KIT for mucosal melanoma. Clinical responses to ipilimumab have been reported in patients with advanced mucosal melanoma with an objective response rate of 6.6% and a median OS 6.4 months.
  4. Ocular melanoma. Ocular melanomas also represent less than 5% of the cases of melanoma. Uveal or choroidal melanomas make up most cases, with conjunctival melanomas occurring less frequently. Specialized ultrasonographic evaluation, together with lesional biopsy, is an important tool in initial diagnosis. Treatment for conjunctival melanoma is complete surgical excision. For localized uveal or choroidal melanomas, there are multiple treatment options and factors such as size of tumor, pathologic diagnosis, and vision in affected eye and contralateral eye, presence of metastasis, patient age, and performance status should be considered. Treatment options include enucleation, radiation, photocoagulation, and thermotherapy. In contrast to conjunctival melanomas, uveal and choroidal melanomas generally metastasize hematogenously to the liver. There is no effective systemic treatment for metastatic ocular melanoma. For patients with liver-dominant metastases, regional therapy should be considered. The MEK inhibitor selumetinib (AZD6244) is currently being evaluated as an investigational agent for metastatic ocular melanoma and preliminary reports confirm single-agent activity (JAMA 2014;311:2397). Ipilimumab has also been studied in metastatic ocular melanoma and preliminary reports document minimal clinical activity.

 Interestingly, the mutation profile of uveal melanoma is distinct from cutaneous melanoma with >85% uveal melanomas showing a GNAQ/GNA11 mutation (most frequently in codon 209); conversely, BRAF mutations are rare. Conjunctival melanoma can infrequently harbor a BRAF mutation. Most major ocular oncology centers now utilize the DecisionDx-UM gene expression profile test (Castle Biosciences) for primary uveal melanoma. This prognostic test (validated in a multicenter clinical trial) stratifies uveal melanoma into three groups based on the 15 gene classifier (Ophthalmology 2012;119:1596). Class 1A patients have a very low risk of metastasis (2% at 5 years), while Class 1B patients have a relatively low risk with 21% developing distant metastasis at 5 years after diagnosis. In contrast, Class 2 patients are at high risk for distant recurrence with 72% developing metastasis at 5 years.

  1. SQUAMOUS CELL CARCINOMA OF THE SKIN
  2. Background
  3. Epidemiology. Squamous cell carcinoma (SCC) is the second most common type of skin cancer in the United States. The overwhelming majority of SCCs occur on chronic sun-exposed skin in older individuals. Men are twice as likely to develop SCC, and its incidence is more than 20 times higher in fair-skinned individuals than in patients with pigmented skin. Incidence also increases with latitudes closer to the equator, reflecting the importance of ultraviolet (UV) exposure in the pathogenesis of SCC.
  4. Risk Factors. The major risk factor for development of SCC is exposure to UV radiation or the sun. Therapeutic sources of UV radiation such as psoralen plus ultraviolet A (PUVA) greatly increase the risk for SCC as do cosmetic sources of UV radiation—indoor tanning accounts for approximately 72,000 excess cases of SCC each year (BMJ 2012;345:e5909). Other risk factors include immunosuppression, especially in the context of solid organ transplant patients, fair skin, exposure to ionizing radiation, infection with certain human papillomavirus subtypes, burn scars, nonhealing ulcers, increased age, and hereditary disorders such as xeroderma pigmentosum or recessive dystrophic epidermolysis bullosa.
  5. Diagnosis and Staging. SCC generally presents as an enlarging, erythematous, scaly papule or plaque on sun-exposed skin that is persistent and may bleed and be tender. It is generally thought to exist on a continuum from precursor lesions, known as actinic keratoses, to SCC in situ (Bowen’s disease) to invasive SCC. Although SCC may be suspected clinically, a biopsy is necessary to make a definitive diagnosis. Several biopsy techniques are adequate, including shave, punch, incisional, or excisional biopsies. Additionally, a full dermatologic examination and palpation of the draining lymph nodes should be performed. In the absence of evidence of metastatic disease, further workup with imaging and laboratory studies is not necessary. Staging of SCCs in the tumor, node, metastasis (TNM) system has been revised in the seventh edition of the AJCC guidelines to include tumor thickness, as it may have prognostic value. One prospective study of SCC in 615 patients demonstrated no metastases in tumors less than 2.0 mm thick, while the rate increased to 4% in tumors 2.1 to 6.0 mm and 16% in tumors larger than 6.0 mm (Lancet Oncol 2008;9:713). However, the existing system tends to cluster poor outcomes to T2 tumors and therefore renders T3 or T4 classifications less meaningful given the rare occurrence of bone metastases. Because of this, an alternative staging scheme has been proposed to better stratify good and poor outcomes.
  6. Therapy. Several treatment options exist for the treatment of SCC. In situ or low-risk lesions in non–hair-bearing locations may be treated with curettage and electrodessication. Most lesions are removed surgically with 0.4-cm margins for lesions smaller than 2 cm in size and more than 0.6 cm margins for lesions larger than 2 cm or with ill-defined borders. Such margins provide cure rates of 90% to 95%. Mohs micrographic surgery can be employed for lesions that are at high risk for recurrence and metastasis, for example, SCC on the central face, ears, eyelids, lips, recurrent tumors, SCCs larger than 2 cm, SCCs with aggressive histologic subtypes, SCCs that develop in scars, or SCCs developing in immunocompromised patients. Mohs micrographic surgery involves the use of frozen or permanent sections to evaluate as close to 100% of the surgical margin as possible by evaluating the circumferential and deep margins. Additional therapies for SCCs include cryosurgery, radiation, and, rarely, intralesional chemo- or immunotherapy. Radiation therapy is generally reserved for patients who are poor surgical candidates. Radiation is used as adjunct therapy in patients with metastatic disease and resected high-risk SCCs, including those with extensive perineural invasion. Patients with advanced disease may benefit from platinum-based combination chemotherapy. Cetuximab (Erbitux, BMS/Lilly), the anti-EGF receptor mAb, has been studied in patients with advanced SCC and shows modest clinical activity (J Clin Oncol2011;29:3419). In metastatic disease, multidisciplinary management and clinical trial are recommended.

There has been some attempt to employ chemoprevention in patients with a high risk of developing cutaneous SCC, particularly solid organ transplant patients. These modalities range from the use of topical immunomodulators, such as imiquimod, to topical and oral retinoids and topical 5-fluorouracil (Efudex). The use of these topical agents results in irritation that is typically well tolerated. Oral retinoid therapy, however, may be associated with serum lipid abnormalities that may already be problematic in this patient population. In addition, the discontinuation of oral retinoid may be associated with a rebound in the number of SCCs.

  1. Prognosis. The vast majority of SCCs can be cured surgically. However, the incidence of local recurrence is 1% to 10%, depending on the method used and can be approximately 20% for high-risk lesions in high-risk locations such as the ear. The incidence of metastasis from cutaneous SCC is 2% to 6%. When SCCs metastasize, they typically go to the first draining lymph node. Certain SCCs have a more aggressive course and are designated as high-risk. High-risk SCCs carry a metastatic risk greater than 10% and include lesions on the lips and ears, lesions larger than 2 cm, thicker lesions, SCCs in scars, recurrent SCCs, SCCs with perineural invasion, and SCCs in immunosuppressed patients.
  2. Follow-up and Prevention. Low-risk SCCs are followed up with full-body skin examinations every 3 to 12 months for the first 2 years, every 6 to 12 months for the next 3 years, and annually thereafter. High-risk SCCs should be followed up with skin and lymph node examinations every 1 to 3 months for the first year, every 2 to 4 months for the next year, then every 4 to 6 months for the next 3 years, then every 6 to 12 months thereafter, according to the 2014 NCCN guidelines. Sun protection and sun avoidance need to be stressed in these patients. In high-risk patients, including solid organ transplant or otherwise immunosuppressed patients, precancerous actinic keratoses should be aggressively treated and threshold for biopsy of suspicious lesions should be low.

 III. BASAL CELL CARCINOMA

  1. Background
  2. Epidemiology. Basal cell carcinoma (BCC) is the most common cancer in the United States, with over 2 million new cases each year. It is more common in men than in women, and its incidence is increasing in all age groups (JAMA 2005;294:681).
  3. Risk Factors. Environmental exposure to UV light from the sun or tanning beds confers significant risk for the development of BCC, both of which are preventable exposures. Approximately 98,000 additional cases of BCC are attributable to indoor tanning (BMJ 2012;345:e5909). Additionally, a history of immunosuppression, increasing age, exposure to ionizing radiation or arsenic, and a history of prior nonmelanoma skin cancer increase risk. Genetic susceptibility to UV damage from fair skin and hereditary disorders such as xeroderma pigmentosum confer risk for BCC. Rare patients will present with numerous, early-onset basal cell carcinomas as a feature of the nevoid basal cell carcinoma syndrome, or Gorlin syndrome, which is due to mutations in the PTCH1 gene. Additional syndromes with early-onset BCC include Bazex–Dupré–Christol and Rombo. Patients with multiple or early BCC or extensive family history should be referred for dermatologic evaluation.
  4. Diagnosis and Staging. BCC classically presents as a pink, pearly papule with a rolled border and arborizing telangiectasias on sun-exposed skin, although common variants include pigmented, ulcerated, or morpheaform morphology. Patients may report that the lesion bleeds easily, does not heal, or is tender. Although in many instances the diagnosis of BCC is strongly suspected based on clinical appearance, biopsy confirms the diagnosis of BCC and provides valuable information to the treating physician. Metatypical, infiltrative, morpheaform, sclerosing, or micronodular features on histology represent an aggressive growth pattern. Any one of several biopsy techniques is acceptable, including shave, punch, incisional, or excisional biopsies. BCCs rarely metastasize, and further workup, beyond a full skin examination, is generally not necessary.
  5. Therapy. BCC is typically treated with destructive or surgical measures. Curettage and electrodessication provides a rapid and effective method to destroy BCCs, with a cure rate more than 90%. Surgical excision of BCCs with at least a 4-mm margin provides a cure rate of approximately 95%. Similar to SCCs, BCCs in high-risk locations, including the “mask areas” of the face and genitals, larger tumors on low-risk areas, tumors with aggressive histology, and recurrent BCCs can be treated with Mohs micrographic surgery. Cryotherapy and radiation therapy are also options for patients with low-risk BCCs who are poor surgical candidates.

For superficial BCC, topical 5-fluorouracil can clear more than 90%, whereas imiquimod has been shown to clear over 80% of such BCCs.

For locally advanced BCC or metastatic BCC, systemic therapy with vismodegib (Erivedge, Genentech) should be considered. Vismodegib is a novel oral small-molecule inhibitor of smoothened homolog (SMO), a downstream target of PTCH1, and has significant clinical activity since most BCCs depend on the activation of the hedgehog (Hh) pathway. In a multicenter phase 2 clinical trial enrolling patients with either locally advanced BCC (n= 63) or metastatic BCC (n = 33), treatment with vismodegib demonstrated significant clinical activity, with objective response rates of 43% in locally advanced disease and 30% in metastatic disease (Sekulic A, et al. NEJM 2012). The recommended dose is 150 mg daily. The most common adverse reactions (≥10%) were muscle spasms, alopecia, dysgeusia, weight loss, fatigue, nausea, diarrhea, decreased appetite, constipation, arthralgias, vomiting, and ageusia.

  1. Prognosis. The prognosis for patients with BCC is excellent. Most patients are cured by the aforementioned modalities. If left untreated, BCCs continue to enlarge and are locally destructive. Metastases occur in less than 0.1% of patients, and common sites are the lymph nodes, lungs, and bones. Although considered incurable, patients with metastatic BCC should be considered for systemic therapy with Hh pathway inhibitor or enrollment in a clinical trial.
  2. Follow-up and Prevention. Patients with a history of BCC have a 50% chance of developing a second BCC within 5 years. Therefore, similar to patients with SCC, close follow-up is recommended with full skin examination every 6 to 12 months. Avoidance of precipitating factors such as sun exposure, tanning beds, and ionizing radiation needs to be stressed in these patients.

 IV. MERKEL CELL CARCINOMA

  1. Background

  1. Epidemiology. Merkel cell carcinoma (MCC) is an uncommon cutaneous cancer. In the United States, approximately 1,500 cases are diagnosed a year. MCC is more common in fair-skinned, elderly individuals, with a mean age of diagnosis between 74 and 76 years of age and 95% of all cases arising in white patients.

  2. Risk Factors and Pathogenesis. MCC is thought to arise from the Merkel cell in the basal layers of the epidermis and has both epithelial and neuroendocrine features. The Merkel cell polyomavirus (MCPyV), discovered in 2008, is thought to play an etiologic role in MCC development and is present in 80% to 100% of examined cases, although the Merkel virus can be found on normal skin or in SCCs. Although the pathogenesis is not clear, there are identifiable risk factors for the development of MCC. These include exposure to the sun and man-made sources of UV radiation, immunosuppression, a history of skin cancer, fair skin, and age more than 70.

  1. Diagnosis and Staging. MCC typically presents as an asymptomatic, rapidly growing, pink-to-red, dome-shaped papulonodule on the head and neck or the upper limbs. Tumors are primarily dermal, although around 10% arise in the epidermis. Biopsy and histologic examination is required to make the diagnosis of MCC, while special stains with a profile of CK-20 and neurofilament positive and TTF-1 negative aid in distinguishing MCC from other neuroendocrine tumors such as small cell carcinoma of the lung (Am J Dermatopathol 2006;28:99). For equivocal lesions, additional neuroendocrine markers should be considered, including chromogranin A, CD56, and synaptophysin. Patients with lesions suspicious for MCC or those who have a confirmed diagnosis of MCC should have a complete history and physical examination, including a full skin and lymph node examination. The seventh edition of the AJCC staging manual uses 2 and 5 cm as size cutoffs for T1, T2, and T3 lesions. Lesions invading the bone, muscle, fascia, or cartilage are considered T4. Stage I disease includes T1 lesions, while stage II includes T2– to T3 lesions without lymph node involvement (stage IIA/B) or T4 (locally invasive primary, stage IIC). Stage III disease is defined by any T lesion with clinical or pathologically apparent nodal disease or in transit metastases, while stage IV is defined by any distant disease. Patients should also have a staging CT, MRI, or preferably PET-CT, which has a specificity of 90% and sensitivity of 90%, to screen for both regional lymph nodes and distant metastases (Am J Clin Dermatol 2013;14:437). PET-CT is particularly useful for evaluation of stage II or stage III disease (tumors larger than 2 cm with no discernable nodal disease). If lymph nodes are negative by clinical examination, it is recommended that patients with stage I or II disease receive a sentinel lymph node biopsy with immunostaining because pathologically negative nodes (stages IA and IIA) confer better outcome than do clinically negative nodes without sentinel biopsy (stages IB and IIB). With clinically positive nodes, fine-needle aspiration should be attempted, and if negative, open biopsy should be considered.
  2. Therapy. The mainstay of therapy for primary disease is wide local excision with 2-cm margins or Mohs micrographic surgery when 2-cm margins are not feasible. Despite removal with wide margins, the chance of local recurrence is high. It is also important to note that the sentinel lymph node biopsy should be done prior to definitive excision, particularly in the head and neck, where the drainage pattern is complex. Although the NCCN guidelines recommend sentinel lymph node biopsies for staging, their effects on survival remain in question. There is some evidence that a wide local excision, combined with adjuvant radiation therapy, may provide a survival benefit. In one retrospective analysis of 1,187 cases from the SEER database, median survival was prolonged with adjuvant RT versus surgery alone (63 vs. 45 months, p = 0.03) (Mojica P, et al. J Clin Oncol2007;25:1043). Patients with recurrent MCC may be treated with a wide local excision, radiation, and chemotherapy if metastatic. Platinum-based chemotherapy with or without etoposide is generally reserved for patients with metastatic disease; however, given the rarity of this tumor, the literature remains sparse regarding systemic therapy for metastatic MCC. For metastatic disease, management by a multidisciplinary management team and enrollment in a clinical trial are recommended.
  3. Prognosis. The 5-year survival rates for MCC vary by stage, with stage IA offering the best prognosis, around 80%, falling to 60% for stages IB and IIA, then around 50% for stages IIB, IIC, and IIIA. Prognosis is poorest for stage IIIB (25% 5-year survival) and metastatic MCC (stage IV, 20% 5-year survival). These numbers are based on the National Cancer Database outcomes from 1986 to 2000. Outcomes tend to be worse for immunocompromised patients.
  4. Follow-up. Given the high rates of local recurrence and metastatic disease, patients should be followed up closely with complete physical examination every 3 to 6 months for the first 2 years, including skin and lymph node examinations. The median time to recurrence is 8 months, with 90% of recurrences in the first 24 months. New approaches to disease monitoring are emerging. Patients with positive oncoprotein antibodies at the time of active disease can track recurrence with a serial examination of their oncoprotein titers (Cancer Res 2010;70:8388).

SUGGESTED READINGS

Gorantla VC, Kirkwood JM. State of melanoma: an historic overview of a field in transition. Hematol Oncol Clin North America 2014;28:415–435.

Eggermont AM, Spatz A, Robert C. Cutaneous melanoma. Lancet 2014;383:816–827.

Griewank KG, Scolyer RA, Thompson JF, et al. Genetic alterations and personalized medicine in melanoma: progress and future prospects. J Natl Cancer Inst 2014;106:435.

Fecher LA, Agarwala SS, Hodi FS, et al. Ipilimumab and its toxicities: a multidisciplinary approach. Oncologist 2013;18:733–743.


Previous
Page
Next
Page

Contents


If you find an error or have any questions, please email us at admin@doctorlib.org. Thank you!