Subungual melanoma (SUM) is an uncommon variant of melanoma that is often difficult to diagnose, both clinically and pathologically. In an attempt to provide pathologic clues to diagnosis, especially in early lesions or small biopsies, and to provide practical advice to pathologists in reporting, the clinicopathologic features of 124 cases of SUM were reviewed, the largest series reported to date. The features of 28 cases of subungual melanoma in situ (MIS), comprising 4 cases of MIS and 24 cases where areas of MIS were present adjacent to dermal-invasive SUMs, were compared with those of a similar number of acral nevi to identify useful distinguishing features. The median age of the patients was 59 years and the most common site was the great toe (24%). Nine percent of cases were AJCC stage 0, 14% were stage I, 41% were stage II, 32% were stage III, and 4% were stage IV at initial diagnosis. The commonest histogenetic subtype was acral lentiginous (66%), followed by nodular (25%) and desmoplastic (7%). The majority of tumors were locally advanced at presentation with 79% being Clark level IV or V. The median Breslow thickness was 3.2 mm. The median mitotic rate was 3 per mm and 33% of cases demonstrated primary tumor ulceration. Seven of 29 patients (24%) who underwent a sentinel lymph node biopsy had nodal disease. Multivariate Cox-regression analysis showed higher disease stage to be the only significant predictor of shortened survival. In comparison to acral nevi, MIS more frequently showed lack of circumscription, a prominent lentiginous growth pattern, predominance of single cells over nests, moderate-to-severe cytologic atypia, a dense and haphazard pagetoid intraepidermal spread of melanocytes, and the presence of junctional/subjunctional lymphocytes ("tumor infiltrating lymphocytes"). Tumor infiltrating lymphocytes have not been highlighted previously as a feature of subungual MIS and represent a useful diagnostic clue. Guidelines for the reporting of SUMs are also presented. Knowledge and recognition of the pathologic features of SUMs and the important features that distinguish them from nevi should reduce the frequency of misdiagnosis.

An FN SN can occur because of deficiencies in nuclear medicine, surgery, or pathology. qRT can detect "occult" metastatic melanoma in SNs that have been identified as negative by histopathology.

These observations suggest a close relationship between the epithelial and stromal elements in phyllodes tumours; endothelin may play a significant role in the malignant progression of phyllodes tumours.

HPV-positive tonsil SCCs have distinct molecular pathways. Their association with younger patient age suggests that they are biologically distinct from HPV-negative tumors.

We appreciate the comments of Tuccari et al regarding our recent study of argyrophilic staining of nucleolar organizer region (AgNOR) counts and morphometry in benign and malignant melanocytic lesions. In this study, we found that both AgNOR counts and AgNOR morphology help to distinguish benign from malignant melanocytic lesions and that the combination of AgNOR counts and morphometric parameters can improve the discriminating ability of the technique. 1 We agree that the AgNOR staining and determination under strictly controlled and preferably uniform conditions will assist in obtaining results that will be comparable among different investigators. We also agree that the use of an image analysis system should help to facilitate this. As pointed out by Tuccari et al and by many others, AgNOR quantification can readily be affected by many factors, including the method of tissue fixation, staining conditions such as temperature and duration, whether an inverted incubation technique is employed, and the methodology applied to determine AgNOR measurements. The bleaching procedure employed in the study of melanocytic lesions is also an important factor that may affect the AgNOR quantification. 2 The application of a standardized AgNOR method, such as that recommended by the Committee on AgNOR Quantitation within the European Society of Pathology, 3 may assist in producing results that are more comparable among different laboratories. However, the presence of melanin pigment within melanocytic lesions will probably limit the usefulness of the methodology recommended by Tuccari et al in assessing AgNOR variables in melanocytic lesions, without the use of a prior melanin bleach procedure such as the one we described. 2 As we have previously demonstrated, 1,2 our procedure has been very successful for assessing both AgNOR counts and AgNOR morphology in melanocytic tumors. It may be that specific standardized AgNOR methods are required for the evaluation of certain tumors. It would thus be valuable if a melanin bleach procedure is incorporated into the standardized AgNOR methods recommended by the Committee on AgNOR Quantitation within the European Society of Pathology 3 for the assessment of melanocytic lesions. Although we agree with Tuccari et al that the AgNOR method has been shown to have prognostic significance in a variety of tumors, 4,5 the purpose of our work was to evaluate the diagnostic utility of AgNOR method in distinguishing benign nevi from malignant melanomas. 1 While we acknowledge that the diagnostic value of individual AgNOR markers may be limited by the phenomenon of overlapping values between benign and malignant lesions, the use of a combination of AgNOR counts and morphometric parameters may improve the diagnostic value of this technique in melanocytic lesions. 1 Our multivariate studies have suggested that the AgNOR diagnostic values in melanocytic lesions may be improved further when co-evaluated with other cytometric markers, such as DNA microdensitometry and MIB1-Ki67 immunoreactivity. 6 Further evaluations of these techniques are required in difficult melanocytic lesions that cannot readily be diagnosed as benign or malignant using standard histopathologic methods. The prognostic significance of the AgNOR technique in melanoma patients using our protocols is yet to be investigated and also requires further study. Ling-Xi L. Li, BM, PhD Richard A. Scolyer, MBBS, FRCPA Kerry A. Crotty, MBBS, BSc(Med), FRCPA Allan A Palmer, MBBS, FRCPA Jillian J. Kril, BSc(Hons), PhD John F. Thompson, MD, FRACS, FACS Stanley W. McCarthy, MBBS, FRCPA

In separate letters the authors discuss the statement that effective management of melanoma is simply about early detection and wide excision. (non-author abstract)

Fine-needle biopsy (FNB) has been reported as a rapid, minimally invasive technique for the diagnosis of metastatic melanoma. The diagnostic accuracy of FNB was assessed in a consecutive series of 2,204 FNBs of clinically suspicious lesions from patients with previous primary melanomas treated at the Sydney Melanoma Unit, Sydney, Australia, between January 1992 and December 2002. The sensitivity and specificity of FNB were 96.3% and 98.9%, respectively. There were 5 false-positive cases (0.6%), which were verified as metastatic adenocarcinoma (3 cases) or reactive processes (organizing hematoma and chronic osteomyelitis, 1 each). False-negative diagnoses (6.7% of cases) were associated with a variety of clinicopathologic factors, including difficult-to-access anatomic sites (eg, high axilla or deep inguinal), small lesions, and lesional characteristics such as fibrosis, necrosis, or cystic change. FNB is a highly accurate, rapid, and cost-effective procedure for the diagnosis of metastatic melanoma and should be considered as the initial diagnostic procedure of choice in patients with melanoma with clinically suspected metastases.

Abrahamsen and colleagues recently reported that more extensive serial sectioning of sentinel lymph nodes (SNs) leads to an increased rate of detection of melanoma micrometastases.1 Although the authors acknowledged that SN analysis involves a compromise between 1) comprehensive sampling and examination of the entire lymph node and 2) practical and economic considerations in routine practice, they did not expound on this issue. We believe that it is important for pathologists and clinicians to understand the practical ramifications of adopting a protocol such as the one used by Abrahamsen et al.1 for routine histopathologic assessment of SNs. If an average time of 1 minute is required to screen each section (a conservative estimate in our experience), and if, on average, 20 sections of each paraffin block obtained from each SN are examined, then, based on median values (2 SNs per patient and 2 paraffin blocks per SN), a pathologist can examine no more than 6 cases in an 8-hour working day! At the Sydney Melanoma Unit (SMU; Royal Prince Alfred Hospital, Camperdown, Australia), we routinely examine four sections of each paraffin block obtained from each available SN; initially, these sections are screened using a 10X objective lens. In 2003, SNs obtained from a total of 434 patients with melanoma were examined, with the examination process consuming approximately 200 hours of pathologists' time. Adoption of the sectioning protocol described by Abrahamsen et al. would call for these pathologists to spend an extra 463 hours examining SNs each year. (Furthermore, additional time would be required for laboratory technicians to cut and stain these sections.) These changes would result in a prohibitive increase in the cost of histopathologic assessment of SNs. We recently reported on the extensive SN biopsy experience accumulated at the SMU.2 Between 1992 and 2001, 26 of 976 patients who had undergone SN biopsy for primary melanoma developed regional lymph node field recurrences after initial pathologic examination of the SNs in these fields yielded negative findings. On the basis of these results, we concluded that efforts to reduce this already very low rate by routinely examining additional sections would be difficult to justify from a cost-benefit perspective. In our view, it is important that practical and economic issues, as well as clinical outcomes, be considered in the development of protocols for detecting metastatic disease in SNs. It is clear that less labor-intensive methods for reliable detection are required, and as we recently reported, nonhistopathologic techniques such as magnetic resonance spectroscopy have the potential to fulfill this need.3

Dendritic fibromyxolipoma (DFML) is an uncommon, recently described, benign soft tissue lesion that shares many clinical and pathological features with myxoid variants of spindle cell lipoma (SCL). As described, DFML is distinguished from SCL by the presence of dendritic cytoplasmic processes, abundant keloidal collagen and a prominent, often plexiform vascular pattern. We describe the first known reported case of an intramuscular DFML that occurred in the right shoulder region of a 73-year-old man. The tumor displayed the typical histopathological features of DFML but also included foci of chondroid metaplasia, a previously unreported finding. This report also discusses the differential diagnosis, particularly distinguishing DFML from SCL and myxoid liposarcoma. In view of the similarities in many clinical and pathological features between SCL and DFML, we speculate that DFML probably represents an unusual variant of myxoid SCL.

Fine-needle biopsy (FNB) has been reported as a rapid, minimally invasive technique for the diagnosis of metastatic melanoma. The diagnostic accuracy of FNB was assessed in a consecutive series of 2,204 FNBs of clinically suspicious lesions from patients with previous primary melanomas treated at the Sydney Melanoma Unit, Sydney, Australia, between January 1992 and December 2002. The sensitivity and specificity of FNB were 96.3% and 98.9%, respectively. There were 5 false-positive cases (0.6%), which were verified as metastatic adenocarcinoma (3 cases) or reactive processes (organizing hematoma and chronic osteomyelitis, 1 each). False-negative diagnoses (6.7% of cases) were associated with a variety of clinicopathologic factors, including difficult-to-access anatomic sites (eg, high axilla or deep inguinal), small lesions, and lesional characteristics such asfibrosis, necrosis, or cystic change. FNB is a highly accurate, rapid, and cost-effective procedure for the diagnosis of metastatic melanoma and should be considered as the initial diagnostic procedure of choice in patients with melanoma with clinically suspected metastases.