Appropriate indication and procedure for random skin biopsy in the diagnosis of intravascular large B‐cell lymphoma

funding source had no role in the study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. The Skin Cancer College of Australasia donated funds to purchase tablets for data entry by patients. [Correction added on 27 October 2020, after first online publication: The Acknowledgements section has been updated.]


Supporting Information
Additional Supporting Information may be found online in Supporting Information: Table S1. Medical Benefits Schedule (MBS) item numbers for the surgical treatment of first primary keratinocyte cancer. Clinically normal samples were obtained from at least three separate areas under local anaesthesia: incisional excision in 45 patients, punch biopsy in 24 cases. Biopsy specimens were stained with anti-CD20 and CD79a antibodies.
A potential association between positive diagnoses and biopsy techniques was evaluated using Fisher Exact Test. Plotting the data showed that neither serum LDH nor sIL-2R levels were normally distributed. Therefore, between-group differences were determined using the Mann-Whitney U test. All analyses were conducted in SPSS (IBM, Japan). Significance was set at P < 0.05.
We diagnosed 10 patients with IVLBCL; the remaining 59 tested negative. One out of 10 IVLBCL-positive cases was detected using punch biopsy, and incisional biopsy detected the rest. Fisher exact test did not detect a significant difference between biopsy techniques.
We did not find significant differences in diagnostic accuracy among biopsy techniques, probably due to the small sample size. However, previous reports recommended against using punch biopsy as less tissue is available and results may not be representative. 4 Incisional biopsy from more than three separate areas, on the other hand, provides more fat tissue and capillaries, and is recommended to avoid false negatives.
Random skin biopsies are usually performed by dermatologists at our hospital. The procedure was mostly used for investigating unknown fevers, pancytopenia and cerebral infarction. For patients with particularly poor health or haematological disorders, we sometimes hesitated to perform incisional random skin biopsies due to possible postoperative haematoma in fat tissues. A previous study indicated five factors (low O 2 saturation, platelet count, serum LDH, serum sIL2R and unexplained fever) that helped to determine the indication for random skin biopsy in patients suspected of having IVLBCL. 2,5 In this study, we focused on serum LDH and sIL-2R because high fever, low O 2 saturation and low platelet count have low specificity, being often due to other complications.
In order to avoid false negatives in cases with serum sIL-2R >2000 U/mL, we recommend incisional biopsy from at least three separate sites. Because in this study, serum sIL-2R levels were over 2000 U/mL for all IVLBCL-positive patients, whereas only 18 out of 59 IVLBCL negative patients were over 2000 U/mL. Conversely, when patients have sIL-2R <500 U/mL and normal serum LDH levels, we do not recommend random skin biopsy because these patients had no lymphoproliferative disorders in our study. Moreover, in cases where patients experience bleeding and coagulation abnormalities, random skin biopsy should be postponed to prevent unexpected serious bleeding. Inter-and intra-patient heterogeneity of PD-L1 expression in metastatic melanomas: A retrospective study Immunotherapy with PD-1/PD-L1 axis blockers by inhibiting the binding of PD-1 with its ligands has become an important strategy in advanced melanoma. There is an objective response in 50% of patients with melanoma metastases, and the expression of PD-L1 is associated with a better outcome, but as a sole biomarker, it can be unreliable. Patients presenting PD-L1-negative metastatic melanomas can also respond to PD-1/PD-L1 axis blockers. 1 B lymphocytes, identified by the presence of CD20 in their membrane, began to be studied more recently in the context of melanomas as a potential biomarker in the monitoring and prognosis of patients. 2 The objective of this retrospective cohort study was to analyse the expression of PD-L1 and CD20 in melanoma metastases, according to site, clinical and histopathological data, and survival. We assessed 50 metastatic melanomas in 46 patients. PD-L1 IHC was performed, 3 utilising E1L3N (R), clone (Cell Signaling Technology -Fig. 1a). We graduated the percentage of PD-L1 in melanoma metastases as negative for cases < 5% of tumour cells, and positive in three subcategories: 5-20%, 21-60%, and >60% of tumours cells expressing PD-L1. For the CD20 study, a similar immunohistochemistry technique was utilised using the clone L26 and the diaminobenzidine chromogen (both from Ventana, AZ, USA, Fig. 1b). CD20-positive cases were subclassified as intratumoral and/or peritumoral. To be classified as positive, CD20 lymphocytes should be intratumoral (within the melanoma bulk) and/or peritumoral (around the melanoma bulk, in close contact with its tumour cells). In the study of lymph nodes, cases in which the only location of the CD20 was in follicular centres (their natural location) were considered negative. Table 1 correlates the histopathological data of metastatic and primary melanomas with the expression of PD-L1 in metastases. Of the 50 cases analysed, 22 (43%) were positive with expression between 5 and 60% of the cells. The age of the individuals ranged from 29 to 96 years (mean = 62 years). The majority were over 50 years old (37/46). The most frequent skin phototypes were II and III. We did not find a statistical difference between the expression of PD-L1 in metastases and the variables 'gender', 'age' or 'phototype'. Patient survival was not associated with the presence of PD-L1. Previous studies assessing survival and PD-L1 expression are contradictory: some relate better prognosis and longer disease-free time, 4,5 and others, a worse prognosis and less relapse-free time. 6,7 We