Sentinel lymph node (SLN) biopsy should be performed for thin melanomas with a tumor thickness of 0.75 mm or greater, suggest the results of a retrospective multicenter studypublished online November 4 in the Journal of Clinical Oncology.
Although the prognostic significance of a positive biopsy is well established with respect to survival, controversy remains as to when the procedure should be performed.
In an attempt to elucidate potential prognostic variables, Dale Han, MD, a surgeon from the Moffitt Cancer Center in Tampa, Florida, and colleagues examined medical records collected by the international Sentinel Lymph Node Working Group from 1994 to 2012. The team identified 1250 patients with melanomas under 1 mm in thickness who had undergone SLN biopsy.
Results showed that most metastases (86.2%) occurred in patients with a Breslow thickness of 0.75 mm or greater, corresponding with an SLN incidence rate of 6.3% (odds ratio [OR], 2.21; 95% confidence interval [CI], 1.06 - 4.61, P = .03). Thinner melanomas fell under the 5% risk threshold for not performing a biopsy, and the overall incidence of SLN disease was 5.2%.
"Given that the morbidity that is associated with sentinel node biopsy remains low in experienced hands, and the finding of a negative sentinel node can save patients the anxiety and expense of intense clinical follow-up, the recommendation…seems reasonable and appropriate," James S. Goydos, MD, from the Rutgers Cancer Institute of New Jersey, writes in an accompanying editorial.
Other significant prognostic factors included ulceration (OR, 2.51; 95% CI, 1.25 - 5.06; P = .01), and a Clark level of 4 or greater (OR, 1.80; 95% CI, 1.01 - 3.23; P = .05). Occult disease occurred in 7.0% and 11.6% of patients, respectively.
Clark levels were also found to have a significant impact on the likelihood of SLN metastasis among patients with melanomas having a thickness of 0.75 mm or greater: 8.2% of those with Clark level ≥ 4 had positive nodal disease compared with 3.7% of those with a Clark level of 3 or less. In contrast, mitotic rates ≥ 1 mm2 were found to have no predictive value, even in patients with tumor thickness ≥ 0.75 mm.
"These findings contradict the findings from other recent studies and the most recent update of the AJCC [American Joint Committee on Cancer] Staging Manual (7th edition) includes mitotic rate in the T classification," Dr. Goydos pointed out.
"It will be interesting to see if future studies confirm Clark level as a significant prognostic variable and whether the next update of the AJCC Staging Manual will once again include this variable in the T classification," Dr. Goydos added.
In an interview with Medscape Medical News, Dr. Goydos remained skeptical about the new findings, especially regarding the importance of Clark levels.
"The problem with Clark levels is that any melanoma over 1 mm thick is almost always Clark level 4 or above and so it is not useful for thicker melanomas. Furthermore, the skin is not a uniform thickness and Clark level 3 in skin of the back or scalp is different from Clark level 3 in skin from the back of the hand or the eyelid. Clark level also doesn’t take into account exophytic growth. The only possible place Clark levels could make a difference is in very thin melanomas, but this is the first modern study showing that it does," Dr. Goydos stated.
With respect to the researchers' findings on mitotic index, Dr. Goydos suggested that phosphohistone H3 might turn out to have better predictive value as a surrogate marker.
"The problem with the current mitotic index used by the AJCC in their staging criteria for melanoma is that it is based on the old-fashioned method of counting mitotic figures — a pathologist looked through a microscope and picked them out by eye," Dr. Goydos pointed out.
"Pathologists can now stain for phosphohistone H3 and the cells undergoing mitosis stand out, making them easy to see. This inflates the number of mitotic figures found in modern specimens. I’ve gotten reports with up to 10 or 12 mitotic figures per high-powered field, something almost unheard of in the past," Dr. Goydos explained.
"The staging systems haven't caught up with the technology yet and so it is not surprising that finding >1 mitotic figure per high-powered field (mm2) wasn't predictive of outcome — there may have been a mixture of old and new methods used to count at the different centers and in different years," Dr. Goydos concluded.
J Clin Oncol. DOI: 10.1200/JCO.2013.50.1114. Abstract
Accompanying editorial. DOI: 10.1200/JCO.2013.51.8423.
Dr. Goydos and Dr. Han have disclosed no relevant financial relationships.
Contributing authors report having received honoraria, renumerations, or compensation for consultant or advisory roles at Merck, Navidea Biopharmaceuticals, and GlaxoSmithKline
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου