WAILEA, HAWAII – The Eighth Edition of the American Joint Committee on Cancer Staging Manual includes significant changes in how melanoma is classified.

The manual has already been published and is available for purchase. However, its implementation will be delayed until Jan. 1, 2018, to give physicians, software vendors, and all other interested parties time to get up to speed. All cancers newly diagnosed through Dec. 31, 2017 should be staged in accord with the seventh edition, released in 2010.

“That’s good news. You have a whole year to become familiar with the changes,” Michael A. Marchetti, MD, observed at the Hawaii Dermatology Seminar provided by Global Academy for Medical Education/Skin Disease Education Foundation.

The eighth edition breaks new ground, moving beyond TNM (Tumor, Node, Metastasis) anatomic staging to incorporate new evidence-based prognostic factors.

“There are some subtle differences here to be aware of. It can be a little bit tricky at first glance. You should become familiar with this,” advised Dr. Marchetti, a dermatologist at Memorial Sloan-Kettering Cancer Center in New York.

In addition to highlighting the changes in melanoma staging included in the new AJCC manual, he outlined key recommendations – some of them controversial – on the use of sentinel lymph node biopsy (SLNB) in melanoma patients incorporated in the 2017 National Comprehensive Cancer Network (NCCN) guidelines.

The biggest change for the dermatology community contained in the new edition of the AJCC staging manual is that the T1 classification of melanoma has changed. In the seventh edition, a melanoma was categorized as T1 if less than or equal to 1.0 mm thickness. The cancer was T1a if nonulcerated and had a mitosis rate of less than 1/mm² and T1b if ulcerated or had at least 1 mitosis/mm².

The eighth edition makes an evidence-based subcategorization of T1 based upon thickness in light of the prognostic implications of this distinction. A melanoma is defined as T1a if nonulcerated and less than 0.8 mm in thickness, and T1b if it is 0.8-1.0 mm thick or less than 0.8 mm with ulceration.

Of note, tumor mitotic rate has been dropped as a staging criterion for T1 tumors.

What this means is, for example, in 2017, a patient with a 0.9-mm nonulcerated melanoma with 1 mitosis/mm² and a negative sentinel lymph node biopsy with wide local excision is T1bN0M0, pathologic Stage IB. Under the eighth edition of AJCC, the same patient is T1bN0M0, pathologic Stage IA, because that mitosis rate isn’t a factor.

Today, a patient with a 0.5-mm melanoma with 1 mitosis/mm² with wide local excision is T1bN0M0, Pathologic Stage IB. Under the new system, the same tumor is downstaged to Pathologic Stage IA, Dr. Marchetti explained.

In the eighth edition, tumor thickness measurements are recorded with rounding to the nearest 0.1 mm, not to the nearest 0.01 mm as before. This change was prompted by the inherent lack of precision in measuring melanomas, especially thicker ones.

The T category definitions of primary tumors have been clarified in the eighth edition. A tumor should be classified as T0 only if there is no evidence of a primary tumor. T is utilized for melanoma in situ. TX is employed when the primary tumor thickness can’t be determined, as for example when the biopsy specimen was obtained through curettage.

The N categorization of regional lymph node status has become much more complicated in the eighth edition, the dermatologist cautioned. Plus, the terminology for nodal disease has changed. The term micrometastasis has been replaced by “clinically occult disease” as detected by SLNB. Macrometastasis has been supplanted by “clinically detected disease.” And while in-transit or satellite node metastasis or microsatellite metastasis with satellite nodes was formerly listed simply as N3, in the new system there are subcategories for N3 based upon the number of metastatic nodes involved. For example, in the eighth edition, a melanoma is pathologic Stage N3a if there are four or more clinically occult regional lymph nodes and no in-transit, satellite, or matted nodes. Pathologic Stage N3b is shorthand for four or more tumor-involved regional lymph nodes, at least one of which was clinically detected, or any number of matted lymph nodes, with no in-transit or satellite nodal involvement. Stage N3c is reserved for melanomas with two or more clinically occult or clinically detected regional lymph nodes and/or any number of matted nodes, plus the presence of in-transit or satellite nodal metastasis.

As a result of the changes in the N classification, there are now four pathologic Stage III groups rather than three. Stages IIIA-C have been joined by pathologic Stage IIID, reserved for patients who are T4b, N3a, b, or c, and M0.

The M categorization of distant metastatic disease status has also become more elaborate. In the AJCC seventh edition, if serum lactate dehydrogenase (LDH) is elevated and a patient has any distant metastatic disease, that’s automatically category M1c. Not any longer, though.

Under the eighth edition, if a patient has distant metastasis to skin, soft tissue including muscle, and/or nonregional lymph nodes and the LDH is unspecified, the categorization is M1a. If serum LDH is not elevated, it’s M1a(0). If elevated, then M1a(1).

Similarly, for distant metastasis to the lung, the range of possibilities based upon LDH is M1b, M1b(0), and M1b(1). For distant metastasis to non-CNS visceral sites, the possibilities are M1c, M1c(0), and M1c(1).

M1d is a new classification, a clear departure from the seventh edition. It applies to patients with distant metastasis to the CNS. The classification is M1d if LDH isn’t recorded, M1d(0) if LDH isn’t elevated, and M1d(1) if it is.

Turning to the updated 2017 NCCN guidelines Version 1.2017 on the role of SLNB in melanoma, Dr. Marchetti noted that the procedure is not recommended in patients with melanoma in situ or Stage IA or IB disease 0.75 mm or less in thickness, regardless of features. Neither are routine imaging or lab tests. That’s because the pretest probability of a positive SLNB is so low, at around 3%.

For Clinicopathologic Stage IA disease, 0.76-1.0 mm in thickness with no ulceration and a mitotic rate of less than 1 per mm², the guidelines recommend that physicians “discuss and consider” SLNB, which the available evidence suggests has roughly a 7% pretest probability of a positive result.

For Stage IB disease, 0.76-1.0 mm in thickness with ulceration or a mitotic rate of at least 1 per mm², as well as for Stage IB or Stage II disease greater than 1.0 mm in thickness, with any feature, the language of the recommendation shifts to “discuss and offer” rather than “discuss and consider” SLNB, since various studies have reported pretest probabilities of a positive result as high as 35%.

“The rationale here for performing sentinel lymph node biopsy is primarily to acquire more staging information. Is it a perfect test? Absolutely not. But it’s the current standard of care in terms of providing additional information for staging,” according to Dr. Marchetti.

If the SLNB generates a positive result, by definition the patient now has Stage III melanoma. The NCCN guidelines recommend consideration of imaging to establish a baseline, and state further that the primary treatment is to discuss and offer complete lymph node dissection in order to control the regional nodal basin and because of a possible favorable impact on overall survival. But the question of a survival benefit has been controversial for many years, and it’s unlikely to be resolved soon, Dr. Marchetti predicted.

The final report from the National Cancer Institute–sponsored Multicenter Selective Lymphadenectomy Trial–1 (MSLT-1) concluded that patients with primary cutaneous melanomas 1.2 mm or more in thickness who were randomized to undergo SLNB and, if positive, immediate complete lymphadenectomy, fared significantly better in terms of 10-year disease-free survival, compared with those assigned to observation and lymphadenectomy in the event of nodal relapse (N Engl J Med. 2014 Feb 13;370[7]:599-609).

This conclusion has generated numerous letters to the editor from melanoma experts who took issue with the analysis and conclusion. To try to put the MSLT-1 results in perspective, Dr. Marchetti applied the results to a hypothetical cohort of 100 patients with intermediate-thickness melanomas of 1.2-3.5 mm undergoing SLNB.

Eighty of these patients would be true SLNB-negative for regional nodal disease. Five others would have a false-negative SLNB and would later develop clinically detectable nodal disease. Fifteen patients with a positive SLNB would undergo prompt complete lymph node dissection, of whom 12 or 13 would derive no mortality benefit at 10 years, assuming the MSLT-1 investigators are correct in their analysis.

“Two or three patients with a positive SLNB will derive mortality benefit at 10 years, but we have no way to identify who those people are from the original 100,” he said.

Since the MSLT-1 report, a phase III German multicenter randomized trial of 241 melanoma patients with a positive screening SLNB has reported results. The participants assigned to complete lymph node dissection didn’t differ in terms of 3-year overall survival, distant metastasis-free survival, or recurrence-free survival, compared with those assigned to observation and lymphadenectomy if nodal disease occurred (Lancet Oncol. 2016 Jun;17[6]:757-67). However, as the investigators noted, the study, known as DeCOG-SLT, was underpowered, and Dr. Marchetti’s view is that it can’t be considered definitive.

“Ultimately I don’t think we’ll have a definitive answer to this question until the final results of the MSLT-II trial in the fall of 2022,” he said.

The MSLT-II trial has the same design as DeCOG-SLT.

Dr. Marchetti reported having no financial conflicts of interest regarding his presentation.

SDEF and this news organization are owned by the same parent company.

bjancin@frontlinemedcom.com

New AJCC guidance brings melanoma staging changes