Demystifying the New Melanoma Staging Guidelines

Of fering a more accurate way of classifying cutaneous melanoma and providing optimal treatment, a new melanoma staging system is making its way onto the scene. The system, the first since 1997, from the American Joint Committee on Cancer (AJCC), was published this May in the AJCC Cancer Staging Manual, Sixth Edition. Official implementation of the new system will take place on Jan. 1, 2003. You need to understand the new tumor-node-metastasis (TNM) system to accurately stage tumors and offer patients the most appropriate course of treatment. This article provides details of the system, highlights major changes, and looks at the implications for your practice. First, let’s review why the new system was put in place. Bring in the New: Why? During the past 5 years, according to Charles M. Balch, M.D., chair of the AJCC Melanoma Staging Committee and professor of surgery and oncology at Johns Hopkins Medical Institutions, Balti-more, MD, substantial advances placed the 1997 system in a position where it wasn’t being used as often as previously. New information needed to be incorporated. “Although a multitude of well-designed single-institution analyses have added to our understanding of prognostic factors in melanoma,” according to an article for which Dr. Balch was lead author in the Aug. 15, 2001, issue of Journal of Clinical Oncology (Vol. 19, No. 16), “few attempts have been made to unify these results into a melanoma staging system to be used in clinical research and clinical practice.” An unprecedented collaborative study was conducted to validate the new system, says Dr. Balch. This involved a database analysis of prognostic factors involving 17,600 patients from 13 cancer centers and organizations, according to the final version of the staging system published in the same issue of Journal of Clinical Oncology. The new evidence-based system should help physicians more accurately stage melanoma. “It’s a better system than the old one in many, many ways,” says Phillip Williford, M.D., associate professor of dermatology and director of dermatologic surgery at Wake Forest University School of Medicine, Winston-Salem, NC. “It defines more accurately those prognostic factors that were shown in large studies to be important.” The new system also holds importance for clinical research. Previously, staging might not have allowed accurate comparisons between studies. “It’s difficult to historically look at trials and make much of them because the staging system was so different,” says Dr. Williford. Having accurate stratification means that a study in France can be validly compared with a study performed in the United States. That could lead to better treatments. Determining the Differences The new system differs in important ways from the previous version. For a side-by-side rundown of differences, see Comparing the 1997 and 2002 staging systems, page 39. Among the most significant changes, as noted by Dr. Balch: • inclusion of ulceration • accounting for stage migration resulting from the use of sentinel node technology • use of tumor thickness as opposed to level of invasion • acknowledgement that the number of lymph node metastases in Stage III is more important than the volume of lymph node metastasis • the understanding that satellite and in-transit metastases are the same biologic entity and that these were artificially split between Stages II and III in the previous system. Ulceration Clinicians agree that including ulceration as part of the staging system represents a significant difference from the previous paradigm. In fact, according to Dr. Balch, it’s the only feature of primary melanoma that can be used to predict outcome in Stages I through III (not Stage IV). “There is pretty clear data from large studies involving tens of thousands of patients,” says Dr. Williford, “that ulceration, as an independent prognostic factor, has a great deal of power in predicting how people do.” In fact, beyond T1 lesions (those greater than 1 mm in thickness), “ulceration probably has the most predictive power — more predictive power than Breslow’s depth,” Dr. Williford says. It also has predictive power with nodal disease, Dr. Williford says he believes. Patients with ulceration and lymph node disease, he says, do worse than those with lymph node disease and no ulceration. He regards nodal disease with ulceration as a “far worse animal” than nodal disease with no ulceration. Ulceration, which is defined histopathologically, not clinically, “pushes you up one notch in terms of a worse prognosis,” says Pearon Lang, M.D., professor of dermatology, otolaryngology and pathology and director of dermatological surgery and oncology, Medical University of South Carolina, Charleston, SC. In the T1 through T4 classification, ulceration moves a patient from an “a” to a “b” category. Traditionally, tumors less than 1 mm were felt to have a good prognosis and required only re-excision, says Arthur Sober, M.D., associate chief of dermatology and director of the melanoma unit, Massachusetts General Hospital, Boston, MA. But under the new staging system, a melanoma less than 1 mm with ulceration might encourage the clinician to consider sentinel node biopsy. “Melanoma ulceration,” says the final version as published in the Journal of Clinical Oncology, “heralds such a high risk for metastases that its presence upstages the prognosis of all such patients, compared with patients who have melanomas of equivalent thickness without ulceration. Thus, survival rates for patients with an ulcerated melanoma are proportionately lower than those of patients with a nonulcerated melanoma of equivalent T category but are remarkably similar to those of patients with a nonulcerated melanoma of the next highest T category.” Stage Migration Because of sentinel node technology, which allows more precise staging of patients, the new system reflects a substantial stage migration, according to Dr. Balch. For instance, patients who previously might have been classified as having Stage II disease based on clinical exam would now be placed into Stage III based on pathologic exam of their lymph nodes. What’s more, says Dr. Balch, clinicians must understand that Stage III represents a diverse group of patients with 5-year survival rates from 13% to 69%. Understanding and using the substages, he says, is quite important. Sentinel node biopsy, says Dr. Sober, allows clinicians to determine whether the patient has metastatic disease to lymph nodes, the most likely location. Finding that disease can upstage the classification to Stage III (regional disease), which changes the direction from observation to potential therapy with immunologically active drugs and surgery. More than thickness, “the sentinel node biopsy is a very powerful prognostic indicator,” according to Dr. Lang. Related to this stage migration is the use of clinical and pathologic staging in the revised system. “A new convention for defining clinical and pathologic staging,” says the sixth edition of the Cancer Staging Manual, “has been developed that takes into account the new staging information gained from intraoperative lymphatic mapping and sentinel node excision.” Clinical staging is based on what clinicians see and feel, versus pathologic staging from a pathologist’s report. Pathologic staging is performed because sentinel node biopsies are frequently done as part of diagnostic and prognostic workups, says Dr. Williford. Clinical staging, says the sixth edition, includes microstaging of the primary melanoma and clinical/radiological evaluation for metastases. Pathologic staging includes microstaging of the primary melanoma and pathologic information about the regional lymph nodes after partial or complete lymphadenectomy. One implication of having clinical and pathologic systems is that a sentinel node biopsy might place a patient into a worse category, reflecting a more accurate real-world assessment of disease. For instance, consider a patient, age 49, with a cutaneous lesion 1.1-mm depth on his right arm who has no clinical evidence of metastatic or node disease. A sentinel node biopsy reveals microscopic involvement in two nodes. For this patient, the clinical stage would be the same but the pathologic stage would be significantly different, says Dr. Williford. What’s more, two groups of these two different types of patients would reveal substantially different outcomes — those with microscopic disease would fare much worse. Thickness Instead of Level of Invasion Both Breslow’s depth and Clark’s level, says Dr. Williford, arose out of an attempt to develop prognostic clues for how patients would fare with presentations of melanoma. But consistently reproducing the Clark’s level was not always guaranteed, suggests Dr. Lang. The new system, which uses thickness measured in even integers, addresses interobserver differences and problems reproducing the Clark’s level, he suggests. Breslow’s depth is more reproducible among different pathologists. “Most pathologists would readily admit that the reproducibility of a Breslow’s depth far exceeds the reproducibility of a Clark’s level,” according to Dr. Williford. “It’s fairly clear,” says Dr. Williford, “that the Breslow’s depth overall is a much more powerful indicator of prognosis than is the Clark’s level.” An overwhelming majority of studies show that adding Clark’s level to Breslow’s depth “adds little or nothing to the prognostic information.” Note, however, that Clark’s level still pertains to the T1 group of lesions. After that, Clark’s level doesn’t contribute to the prognostic model. Dr. Williford suggests that some would find this an anachronism that should be eliminated. After examining the primary melanoma, the pathologist needs to provide the dermatologist with three items, says Douglas S. Reintgen, M.D., cancer center director, Lakeland Regional Cancer Center, Lakeland, FL: tumor thickness, Clark’s level and ulceration. Of these, tumor thickness and ulceration are likely most important, he suggests. Clark’s level becomes important for melanomas that are less than 1-mm thick. Volume of Lymph Node Metastasis While the previous system accounted for size of lymph nodes, which reflects volume, the number of lymph nodes involved in metastasis is far more important, says Dr. Lang. The new system reflects this change. “The number of metastatic lymph nodes, rather than their gross dimensions and the delineation of clinical occult (i.e., ‘microscopic’) vs. clinically apparent (i.e., ‘macroscopic’) nodal metastases, are used in the N category,” states the sixth edition of the Cancer Staging Manual. Dr. Reintgen suggests that clinicians weren’t certain how the volume of lymph node metastasis appeared in the previous staging system. Dr. Williford says he’s not aware of one study that shows that size of a lymph node is important in how patients fare, while many studies show that the number of lymph nodes is important. Satellite and In-Transit Metastases Both satellite and in-transit metastases have broken off the main tumor, lodged en route to lymph nodes, and grown, says Dr. Lang. In his experience, some patients with satellite lesions can survive for years. Application in Practice With this revised, more complex melanoma staging system, you’re faced with applying the method in practice. Experts offer various points to keep in mind. • Remember that Stage I and II melanoma are confined to patients with localized melanoma, defined either clinically or pathologically, says Dr. Balch. Stage III is confined to regional metastasis, including satellite, in-transit, and lymph node metastasis. For all stages, clinicians are strongly encouraged, Dr. Balch says, to indicate clinical or pathological staging. • Patients with T1B, T2, T3 and T4 melanoma who are clinically node negative should be considered for a sentinel node exam as part of pathological staging, especially those who are potential candidates for melanoma clinical trials, says Dr. Balch. This is a formal recommendation of the AJCC Melanoma Staging Committee. • The new system, says Dr. Williford, acknowledges that, as with most solid tissue tumors, the finding of cancer in a lymph node system or equivalent is a poor prognostic sign. The system thus allows clinicians to better stage patients who heretofore clinically would have been staged as having not that serious a disease. This allows sentinel node biopsy to substantially improve the ability to stage patients, define a prognosis, and put them in various treatment groups to see if the physician can find a treatment to improve overall outcome. • If your patient’s melanoma thickness is 1.0 mm or greater, discuss with him or her possible referral for a sentinel node biopsy for prognostic reasons, Dr. Williford says. • For patients who have a defined risk of nodal disease, make sure to refer the patient for lymphatic mapping at 1-mm depth, Dr. Reintgen notes. • Don’t lump all Stage III patients into one basket of poor prognosis, notes Dr. Williford. Outcomes vary significantly between Stages IIIA, IIIB and IIIC. • Biopsying a suspicious lesion to obtain tumor thickness is crucial, says Dr. Reintgen. Make sure you get entirely beneath the lesion. • Avoid performing a wide re-excision before the sentinel node biopsy, according to Dr. Sober. The sentinel node biopsy will then be more valid and predictable. Potential Concerns Though by virtually all accounts the revised melanoma staging system represents a substantial advance, the new paradigm isn’t devoid of concerns. In a letter in the April 15, 2002, issue of Cancer, a group of German writers suggests that ulceration doesn’t appear to have a prognostic impact for all the different classes of tumor thickness. Upstaging based on ulceration, they suggest, should be taken into account only for T2 and T3 melanomas. They also hold that the staging system should remove the level of invasion from the T1 tumor subgroup. The writers also suggest removing ulceration of the primary tumor from the N classification. Also, they say that satellite and in-transit metastasis would be better assigned to the N1 category instead of N2. Finally, they recommend eliminating the lactate dehydrogenase value in the M category. These items were considered in the committee’s decision-making, says Dr. Balch. Another issue involves the definition of ulceration. An article in Journal of Clinical Oncology states that “the interpretation of melanoma ulceration among pathologists is one of the most reproducible of all the major histopathologic features.” But Dr. Williford holds that ulceration isn’t extremely well-defined. In his view, ulceration should involve full thickness loss of epidermis and dermis. Among pathologists, he feels there needs to be a better definition of what constitutes a histopathologic diagnosis of ulceration. Some pathologists mistakenly classify erosions as ulcers, he suggests. Future Phases The Journal of Clinical Oncology reports that the next phase of staging melanoma will evolve as new technology allows diagnosing at a level of tumor burden better than that achievable with the light microscope or routine X-rays. Advances such as polymerase chain reaction and DNA microarray technology will likely find their way into the melanoma staging system, says Dr. Balch (see Melanoma and DNA microarrays, page 47.). For today, however, the new melanoma staging system holds promise for better categorization and treatment of this increasingly prevalent disease. Understanding and using the system accurately can help you achieve the best outcome for your patients.