Melanoma (January 2017): Molecular Oncology Tumor Board

ASCO University
Jan 11, 2017 2:00 PM

User Instructions: Welcome to the Molecular Oncology Tumor Board Series! This educational initiative is a collaboration between the American Society of Clinical Oncology (ASCO), College of American Pathologists (CAP), and Association for Molecular Pathology (AMP).

 

A new case will be presented each month with discussions led by an expert pathologist and medical oncologist. Have an interesting case in mind? Submit your hypothetical patient cases for consideration in an upcoming Molecular Oncology Tumor Board discussion forum.

 

This month’s topic is led by Drs. Ryan Sullivan (Medical Oncologist, Massachusetts General Hospital) and Allison Cushman-Vokoun (Pathologist, University of Nebraska Medical Center).

 

Users are encouraged to leave comments and post questions about the case in order to generate a wide discussion among the cancer care community. You can also receive email notifications when new comments are posted by clicking the “Follow this Conversation” option located at the bottom of this page.

 

When posting, please abide by the terms and conditions of this website.

 

 

 

 

Comments

13846

ASCO University
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 11, 2017 2:08 PM

Patient Case

Patient History: A 41-year-old man noted a dark area on the roof of his mouth in the Spring of 2015, in addition to areas on his left upper gums several months earlier.  At the urging of his wife, he scheduled an appointment with a dentist for the first time in years, and prior to that appointment, his wife observed a new mass in his left neck.  He was seen by an otolaryngologist, who performed a biopsy of the lesion (picture #1) which showed melanoma.

Type of Tumor/Pathology: Specifically, the pathology showed an at least 3.0 mm thick, ulcerated melanoma of unclassified histologic subtype, with 3 mitoses per square millimeter. A targeted next generation sequencing assay was performed on the tumor and identified two genetic abnormalities:

Single nucleotide variants:
SNV 1 of 1: BRAF ENSP00000288602:p.Val600Glu (ENST00000288602.6:c.1799T>A)

Insertions/deletions:
Indel 1 of 1: CDKN2A ENSP00000462950:p.Pro72fs (ENST00000579755.1:c.215delC)

Images/Scans: Staging MRI brain and PET/CT imaging showed no evidence of brain metastases, but demonstrated intense FDG uptake in the region of the left hard palate with extension into the alveolar ridge of the left maxilla, consistent with a neoplastic process. A midline focus of FDG accumulation in the soft palate, as well as several level I and II lymph nodes in the left neck, were highly suspicious for neoplastic involvement. The patient also had multiple mediastinal and hilar lymph nodes that measured up to 5 mm and did not demonstrate FDG avidity.

13851

ASCO University
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 11, 2017 2:15 PM

Discussion Questions

  1. Which tests are the most appropriate in identifying a BRAF codon 600 mutation for potential FDA-targeted treatment?

  2. What is the significance of the CDKN2A mutation?

  3. If systemic therapy is selected, would you be more inclined to offer targeted therapy or immune therapy for this patient?

 

13856

Anis Toumeh, MD
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 12, 2017 10:10 AM

The presentation is of mucosal melanoma, which is an aggressive type with infrequent senstivie BRAF mutations. This patient definitley needs multidiciplinary discussion including surgical consultation in case of possible resection including the neck nodes. The hilar and mediastinal nodes probably need to be sampled if there is a surgical consideration. The location may preclude resection with negative margins. Emerging data regarding the efficacy of immunotherapy with single agent Nivolumab and combination of ipilimumab and nivolumab (D'Angelo et al. JCO DOI: 10.1200/JCO.2016.67.9258)

The presence of CDKN2A mutation might suggest familial melanoma.

Ryan J. Sullivan, MD
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 18, 2017 8:02 PM

These are excellent points.  

The data provided in the reference paper suggests that this anatomical subset of patients have responses to anti-PD1 based therapy; albeit at a lower rate than patients with non-acral, cutaneous melanoma.  

Germline CDKN2A mutations are associated with familial melanoma.  However, mutations, deletions, and hypomethylation of CDKN2A occurs in at least half of cutaneous melanomas (Melanoma TCGA, Cell - reference #2), whereas patients with germline CDKN2A mutant patients likely represent much less than 1% of all patients with metastatic, cutaneous melanoma.  Thus, in a patient without a strong family history or a personal history of multiple primary melanomas, genetic counseling generally is not recommended.

13866

Allison Cushman-Vokoun
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 16, 2017 12:19 PM

Course Faculty Response: Dr. Cushman-Vokoun, MD, PhD

There are two FDA-approved assays for BRAF mutation testing in melanoma, both of which are real-time PCR-based and were approved as companion diagnostic tests.  These include the BioMérieux THxID®-BRAF, which detects V600E and V600K and the cobas® 4800 BRAF V600 Mutation Test, which states that it detects V600E and may detect others such as V600K and V600D.    Real-time PCR has a high analytical sensitivity (down to <1% mutant allele), so it can be beneficial in low tumor specimens (<10% tumor). However, in the case of melanoma, usually tumor burden isn’t a problem. Conversely, real-time PCR can only detect mutations for which the primers and probes are designed and this should be kept in mind as there are multiple BRAF variants in melanoma both in codons 600, 601 and others in exons 11 and 15.  While there are only FDA-approved drugs for p.V600E and p.V600K mutated tumors, other mutations in BRAF may also be treatable by BRAF inhibitors, and some BRAF mutations actually are inactivating (e.g. D594 mutations) to the protein and may not be good targets for current drugs.  

Many labs employ other methods than real-time PCR in order to identify more mutations.  Each of these have pros and cons. Dideoxy (Sanger) sequencing can cover a whole exon in one tube (such as BRAF exon 15, which includes codons 594, 597, 600 and 601), but requires more tumor percentage (30-40%) due to its low analytical sensitivity of 15-20% mutant allele (unless PCR enrichment strategies are used). Pyrosequencing has a higher analytical sensitivity (5%) but usually only includes a couple of codons per tube (ideal for codon 600 mutations). Many labs are now performing melanoma panels by next generation sequencing (NGS) to simultaneously interrogate multiple genes mutated in melanoma or mimickers such as Spitz nevi (e.g. BRAF, NRAS, KIT, GNAQ, GNA11, HRAS) to assist in diagnosis, therapy, or clinical trial selection.  NGS has a high analytical sensitivity (1-5%) and can assess multiple regions in multiple genes. It requires more technological and bioinformatics sophistication, but is becoming the cheapest and quickest way to assess melanoma mutational profiles. NGS is also being used more to analyze mutation burden of tumors, in order to determine response to immunomodulatory therapy. All of these techniques can be done on DNA extracted from formalin-fixed paraffin-embedded specimens, but one must remember that melanin can inhibit PCR. This is usually removed by extraction, but sometimes bovine serum albumin is required to bind melanin and prevent it from inhibiting a PCR reaction. Each technique described above has pros and cons, which a lab must consider before employing (Table 1).

 

13871

Ryan J. Sullivan, MD
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 16, 2017 12:27 PM

Course Faculty Response: Dr. Sullivan, MD

CDKN2A is a gene that encodes for two proteins, P14ARF and P16INK4A, that function to regulate the cell cycle by acting as cyclin dependent kinase (CDK) inhibitors. (1) When this gene is functionally lost through an inactivating mutation, such as a frame shift deletion, the cell cycle

regulation of G1 to S phase is lost and proliferation is promoted. (1) In melanoma, CDKN2A aberrations are common, have been shown to be a mechanism of acquired BRAF inhibitor resistance, and are associated with reduced benefit to BRAF-targeted therapy when present at baseline. (2-5) There is emerging data that dual targeting of BRAF and CDK4/6, which is activated with loss of function of CDKN2A mutations, may be a useful therapeutic strategy. (6, 7)

The decision of frontline therapy for a patient with unresectable or metastatic BRAFV600 –mutant melanoma has become very complicated.  With the approval of eight drugs for this patient population in the past six years, there are many options but very little data about predicting which patients are best served with an upfront BRAF-targeted therapy or immune-checkpoint inhibitor approach.  In general, the selection of therapy is by gestalt and is not based on prospective evidence. For example, most melanoma specialists consider offering BRAF-targeted therapy for patients with rapidly progressing or highly symptomatic disease due to the high response rate associated with combined BRAF/MEK inhibitor combination therapy. Conversely, patients with lower volume or asymptomatic disease are offered an anti-PD1 inhibitor (pembrolizumab, nivolumab) either as a single agent or in combination with the anti-CTLA4 antibody ipilimumab. 

With regards to BRAF-targeted therapy, the frontline data is very straightforward.  Single-agent BRAF inhibitor (vemurafenib, dabrafenib) and MEK inhibitor (trametinib) therapy are associated with improved outcomes compared to chemotherapy. (8-10) This data led to the approval of single-agent vemurafenib (2011), dabrafenib (2013), and trametinib (2013). Further, combined BRAF and MEK inhibitor therapy is associated with improved outcomes compared to single-agent BRAF inhibitor therapy without a major increase in toxicity. This finding has been demonstrated in four trials (COMBI-d, COMBI-v, coBRIM, COLUMBUS) with three different combinations (dabrafenib & trametinib, vemurafenib & cobimetinib, and encorafenib & binimetinib), and to date, has led to the approval of two of these combinations, dabrafenib & trametinib in 2014 and vemurafenib & cobimetinib in 2015.  (11-14) The longest follow up has been with dabrafenib & trametinib, showing responses in 69% of patients with a median progression free survival (PFS) 11.0 months, a three-year PFS of 22%, median overall survival (OS) of 25 months, and a three-year OS of 44%. (15) From this same dataset, a post-hoc analysis has been performed showing that the survival data was even better when looking at patients with a normal LDH and less than three disease sites: three-year PFS 38% and three-year OS 62%. 

The frontline data with immune checkpoint inhibitor therapy is evolving, but a general summary is as follows. First, therapy with ipilimumab is associated with improved outcomes in two randomized phase III trials that led to its FDA-approval in 2011. (16, 17) Furthermore, the long term outcome data suggests a plateau of the overall survival at approximately 20%. (18) Nivolumab and pembrolizumab, which have been shown to be superior to ipilimumab and were both FDA-approved in 2014, are associated with responses in 35-45% of patients in the front line setting, and a median OS of approximately 24 months. (19, 20) The combination of ipilimumab and nivolumab is associated with higher responses (in excess of 60%) and improved PFS, than nivolumab; however, is also associated with a much higher toxicity rate (Grade 3/4/5 toxicity of 60% versus 10-15% with nivolumab or pembrolizumab) that limits its widespread applicability. (21) Additionally, there is no overall survival data available to date, making it difficult to understand the value of upfront combination therapy versus sequencing of therapy.

A summary of the frontline data is found in this table. Unfortunately, there is no prospective data to help guide treatment decision making for BRAF mutant melanoma patients; although, a large cooperative group trial (EA6134; NCT02224781) is actively enrolling to provide a definitive answer to whether frontline combined immune checkpoint inhibitor therapy compared with combined BRAF/MEK inhibitor therapy is superior.  However, until the results of this trial are available, providers are left only with retrospective evidence and post-hoc analyses from large prospective studies to help make decisions for their patients. 

See references here.

13876

ASCO University
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 16, 2017 12:37 PM

Patient Case Update

After discussion with the patient and his family, a decision was made to commence systemic therapy in place of surgery for two reasons.  Firstly, this patient had at least Stage IIIC melanoma and quite possible stage IV (prominent if not pathologically enlarged mediastinal and hilar adenopathy); both associated with a very high risk of death at five years (exceeding 80% and 90% respectively).  In the setting of such a high risk, performing a highly morbid reconstructive therapy was thought to be unwise. Secondly, there was an expectation that the patient would respond to therapy given the high response rate associated with combination checkpoint inhibitor therapy as well as combination BRAF and MEK inhibitor therapy. Further, if he had a complete response to immunotherapy, he may never need to have surgery, and if he did not, then it may be possible that surgery would be less morbid.

Due to the emerging response and PFS data with nivolumab and ipilimumab and its recent FDA approval, the patient commenced therapy and completed three of four induction doses.  His third dose was held due to fevers and malaise, but he received his fourth without major toxicity and transitioned to nivolumab maintenance after his imaging demonstrated evidence of a partial response to therapy.  This has been continued for over a year, and the patient continues to tolerated therapy well and have a partial response to therapy. 

13881

ASCO University
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 16, 2017 12:41 PM

Discussion Questions

1. What is known about the pathological and molecular characterization of mucosal melanoma?

2. Does location of the primary tumor influence response to therapy?

3. Does tumor mutational status associated with response to immune therapy?

 

13891

Allison Cushman-Vokoun
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 20, 2017 9:45 AM

Course Faculty Response: Dr. Cushman-Vokoun, MD, PhD

Mucosal melanomas arise from melanocytes within the oral, nasopharyngeal, gastrointestinal, respiratory or ano-genital tracts. Mucosal melanocyte function is not completely understood; however, the cells are likely involved in innate immunity. Unlike cutaneous melanomas, which arise from UV exposure, the underlying cause of the development of mucosal melanoma has not been determined. They are associated with increasing age (65% older than 60 years old) and have a higher incidence in the Caucasian population (2X higher). Mucosal melanomas can behave more aggressively and have a poorer prognosis, probably in part because they are more difficult to detect.  

Mucosal melanomas have a somewhat different genetic profile than cutaneous melanomas in that KIT alterations (amplifications or mutations) are more common (in up to 40% depending on the study; 15-20% are mutations). It is important to identify KIT mutations because some mutations can be responsive to imatinib therapy.  NRAS mutations can also be found in mucosal melanomas (15%), but BRAF mutations are rarer in mucosal melanomas (5-10%). That being said, BRAF testing should not be withheld in mucosal melanomas, as illustrated by this case. 

See references here.

13896

Ryan J. Sullivan, MD
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 20, 2017 9:48 AM

Course Faculty Response: Dr. Sullivan, MD

Melanoma is a malignancy of the pigment producing cells in the body, melanocytes, and may arise in the skin, structures of the eye, and in mucosal surfaces. The clinical data of immune checkpoint inhibitor use across various anatomic subtypes of melanoma shows that patients with cutaneous melanoma have a higher response rate with single-agent anti-PD1 inhibitors and the combination of ipilimumab and nivolumab, than those with mucosal melanoma where the response rates are nearly half that of cutaneous melanoma (22). Further, patients with uveal melanoma have response rates to checkpoint inhibitor therapy that tends to be approximately 5% (23, 24). To date, there is not enough publically available data to comment on the rate of response to combined therapy with ipilimumab and nivolumab.

The driving etiologic factor in cutaneous melanoma is ultraviolet (UV) irradiation exposure from the sun or tanning that causes cancer-causing and non-cancer causing mutations; however, this does not appear to be associated with the development of mucosal or uveal melanoma. As such, the total mutational burden is, as expected, to be higher in patients with cutaneous melanoma than in mucosal melanoma patients, which is higher in patients with uveal melanoma (25). In numerous reports, mutational burden has been associated with increased responsiveness to immune checkpoint inhibitor therapy; thus, it is possible that the differences in mutational burden across anatomic subtypes is, at least in part, responsible for the differences in responsiveness to this type of therapy (26).

A recent analysis of the cutaneous melanoma cancer genome atlas (TCGA) identified four main molecular subtypes of melanoma based on the dominant melanoma oncogene: BRAF (~50%), NRAS (~25%), NF1 (~10%), triple-wild type (~15%) (2). BRAF mutational status has been shown to be associated with similar responsiveness to high-dose IL2, ipilimumab, and likely anti-PD1 therapy (single-agent and combinations).  NRAS mutations have been demonstrated to be associated with increased responsiveness to immunotherapy in some series, but not others. Therefore, it is probably safer to say that it is inconclusive as to whether NRAS mutations are associated with better responses to immunotherapy (26, 27, 28. NF1 mutations, which have a significantly higher mutational burden than the other subtypes, are associated with a higher response to anti-PD1 therapy (2, 26). This is presumably due to the fact that these lesions tend to be located on sun-exposed areas and have higher mutational burdens than tumors in other molecular subtypes.

See references here.

Ahmed Tarig Ahmed, MBBS
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Feb 28, 2017 9:59 PM

I have a patient with metastatic melanoma with axillary and supra-clav adenopathy. Asymptomatic otherwise. Would you consider immunotherapy first in this case vs BRAF/MEK inhibitors? 

13906

Ryan J. Sullivan, MD
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 25, 2017 12:13 PM

Course Faculty Summary: Dr. Sullivan, MD

  • The treatment for BRAF mutant melanoma has been revolutionized over the past five years and both BRAF-targeted and immune therapy regimens are reasonable front-line options.
  • Additional genetic alterations, such as CDKN2A aberrations or total mutational burden, may influence the responsiveness to BRAF-targeted and immune checkpoint inhibitor therapy.
  • Immune checkpoint inhibitor therapy responsiveness varies in melanoma according to the anatomic site of primary melanoma (cutaneous > mucosal > acral > uveal).

13911

Allison Cushman-Vokoun
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 25, 2017 12:18 PM

Course Faculty Summary: Dr. Cushman-Vokoun, MD, PhD

  • There are multiple BRAF variants found in melanoma; currently two (V600E and V600K) are included in FDA approved drug labels.
  • When evaluating for BRAF mutations, labs should consider their techniques and resources available and assess pros and cons for each assay prior to instituting.  Oncologists should be aware of strengths and weaknesses of the assay in use, the limit of detection of the assay, and tumor requirements.
  • BRAF is not the only gene mutated in melanoma that is targetable and Melanoma Panels should be considered, especially if a melanoma is BRAF negative. Next generation sequencing (NGS) is making these easier and cheaper to do.

13916

ASCO University
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Jan 25, 2017 12:20 PM

Thank you to Drs. Cushman-Vokoun and Sullivan for leading the discussion of this case and also to all of those who contributed to the conversation! The forum is now closed to further comments and participants can claim CME credit and 1 ABIM MOC point through ASCO University.

Please check back in mid-February for a new case in this series related to chronic myelogenous leukemia (CML).    

Have an interesting case in mind? Submit your hypothetical patient cases for consideration in an upcoming Molecular Oncology Tumor Board discussion forum.

14226

Ryan J. Sullivan, MD
Re: Melanoma (January 2017): Molecular Oncology Tumor Board
Mar 02, 2017 9:46 AM

Response to question posted by Ahmed Tarig Ahmed, MBBS on Feb 28, 2017:

Assuming this patient has a BRAFV600 mutation, there are a number of potential treatment options including immunotherapy and BRAF targeted therapy.  If there is no other disease and the patient has borderline resectable disease or otherwise disease that may become resectable with a marked response to therapy, then I would strongly consider BRAF/MEK inhibitor combination therapy, assuming also no contraindications to therapy, and then surgery if the patient’s disease becomes resectable.  After the patient has been rendered disease free with surgery, it would be reasonable to discontinue therapy and watch expectantly.  If, however, there is no real role for surgical resection or if other areas of metastatic disease are present, then our practice is typically to use upfront anti-PD1 therapy as a single-agent; particularly if the disease is not rapidly progression or symptomatic.