© 2014 The Authors APMIS © 2014 APMIS DOI 10.1111/apm.12263
APMIS 122 (Suppl. 137): 1–17
O1-1 Histological deﬁnition of a lymph node - based on a survey among Danish pathologists Agnete Riisgaard Sørensen1, Astrid Petersen1 1 Department of Pathology, Aalborg University Hospital, Aalborg, Denmark Introduction: Harvesting a speciﬁc number of lymph nodes for pathological examination of surgical cancer specimens is recommended in national and international guidelines including the TNM classiﬁcation. The number of lymph nodes with metastasis is also a parameter for the diﬀerent N-stages in some types of cancer. In contrast the exact minimal histological criteria for a lymph node are not clearly deﬁned in the literature. We have conducted an investigation on which deﬁnition is used by Danish pathologists and residents in pathology. Material and methods: A questionnaire in Survey Monkeyâ was sent by email to members of the Danish Society of Pathology on December 11th 2013. The responses were collected on January 6th 2014. The questionnaire oﬀered four diﬀerent options for the histological criteria for a lymph node, three options given as a yes/no answer and one option as free text. A ﬁnal question of the position of the responder was also given. Results: 261 members of the Danish Pathology Society received the questionnaire and 138 (53%) answered. 99 (72%) were specialists in pathology, 34 (25%) were resident doctors in pathology, 3 (2%) designated their position as ‘other’ and 2 (1%) did not answer the question of their position. 47 (34%) had as the minimal requirement for a lymph node a nodular collection of lymphatic tissue with both a capsule and a sinus. Of these 35 (74%) were specialists, 10 (21%) residents, 1 (2%) ‘other’ and 1 (2%) had not given their position. 29 (21%) required a nodular collection of lymphatic tissue with either a capsule or a sinus. Of these 21 (72%) were specialists, 6 (21%) residents and 2 (7%) ‘other’. 38 (28%) required a nodular collection of lymphatic tissue with a capsule. Of these 25 (66%) were specialists, 12 (32%) residents and 1 (3%) had not given their position. 21 (15%) required a nodular collection of lymphatic tissue with a sinus. Of these 16 (76%) were specialists and 5 (24%) residents. 3 (2%) had a personal deﬁnition (a bigger collection of lymphocytes (1), a lymphocytic collection with a capsule and lymph node architecture (2)). All these were specialists. Discussion and conclusion: It seems clear from our questionnaire that the pathologists in Denmark do not have a consistent deﬁnition of what a lymph node is. This missing uniformity may invalidate the comparison between diﬀerent pathology departments e.g. in the number of identiﬁed lymph nodes. A consensus upon a common deﬁnition is advisable.
O1-2 Processing times for gastrointestinal biopsies with suspected malignancy can be reduced by a simple intervention Tina Prangsgaard1,2, Ulla Engel1 1 Hvidovre University Hospital, Department of Pathology, Hvidovre, Denmark; 2Department of Pathology, Herlev Hospital, Denmark Introduction: By 2009 National Integrated Cancer Pathways (ICP) were implemented in the Danish health care system. ICP are organizational and clinical standards for the diagnostics and treatment of cancer in Denmark. If malignancy is suspected, the examining doctor can prioritize samples as ICPspecimens - opposed to routine samples. Guidelines recommend that 90% of pathology reports from ICP should be issued in 3 days. The purpose of ICP is to accelerate the diagnostic process and onset of treatment. The aim for the pathology department is to reduce processing-times through isolation of ICP-specimens. This study intended to evaluate if samples, in which the biopsy-taking doctor suspected malignancy, were correctly prioritized as ICP-specimens. Furthermore we wanted to examine if a simple intervention could improve the proportion correctly prioritized as ICP-specimens and thereby reduce processing-times for biopsies suitable for an ICP. Material and methods: All biopsies from the gastrointestinal tract, received at the department in May and June 2013 were included and classiﬁed into three groups: A: Biopsies in which malignancy was suspected prioritized correctly as ICP-specimens. B: Biopsies in which malignancy was suspected prioritized incorrectly as routine specimens. C: Biopsies in which malignancy was not suspected prioritized correctly as routine specimens. Intervention: Doctors who sent biopsies categorized as B in May 2013 were contacted by telephone and informed about the possibility and importance of correct prioritizing of ICP specimens. Results: 946 biopsies were received in May 2013 prior to the intervention. 53 samples were suitable for an ICP (A+B). 57% (n = 30) were correctly classiﬁed as ICP (A) and 43% (n = 23) were wrongly classiﬁed as routine samples (B). 75% (n = 40) of the samples suitable for an ICP were assigned with a diagnosis within 3 days. 909 biopsies were received in June 2013 following the intervention. 63 samples were suitable for an ICP (A+B). 84% (n = 53) were correctly classiﬁed as ICP (A) and 16% (n = 10) were wrongly classiﬁed as routine samples (B). 90% (n = 57) of the samples suitable for an ICP were assigned with a diagnosis within 3 days. The most frequent reasons for incorrect prioritizing of the biopsies were forgetfulness or unawareness of the possibility of ICP prioritizing. Discussion and conclusion: Awareness and use of ICP among biopsy-taking doctors are necessary to secure optimal conditions for the diagnostics and treatment of cancer. Wrong
prioritizing delays the diagnostic process. We showed, that a telephone call to doctors who prioritized gastrointestinal pathology specimens wrongly, increased the proportion of biopsies prioritized correctly and decreased response times for ICP-specimens.
O1-3 Validation of automated image analysis on HER2 expression in breast cancer Henrik Holm Rossing1, Martin Kristensson2, Maj-Lis Møller Talman1, Ben Vainer1 1 Department of Pathology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark; 2Visiopharm, Hoersholm, Denmark Introduction: Human epidermal growth factor receptor 2 (HER2) is a receptor for circulating growth factor, stimulating uncontrolled cell proliferation. The trastuzumab (Herceptin) antibody reacts with HER2, arresting the cells in the G1 phase and presumably also inducing an immune system-directed cell killing. In breast cancer, overexpression of HER2 occurs in 15–20%, and analysis of HER2 expression is therefore pivotal for selecting the correct patients for this treatment. The aim of the study was to validate the digital, automated image analysis algorithm HER2-CONNECTTM, which is developed to discriminate between ampliﬁed and non-ampliﬁed HER2 gene expression, with the ﬁnal goal of minimizing the number of inconclusive 2+ scores. The validation is part of a general implementation of automated image analysis for routine breast cancer evaluation. Materials and methods: Consecutive samples (n = 315, from 157 patients) received at the department within the period week 23–34 were included. TMAs were routinely manufactured, and each core was reviewed in order to ensure the presence only of invasive carcinoma. Immunostaining was performed with Dako’s ready-to-use HER2 test (Glostrup, Denmark). TMAs were scanned in a Mirax Midi Scanner (3D-Histech), and one batch analysis of the HER2-CONNECT algorithm was run including all samples. This algorithm evaluates the immunohistochemical staining reaction of HER2 based on cell membrane connectivity by using a dynamic measure (between 0 and 1) of the size of HER2stained membrane fragments. The automatically read connectivity translates into the classic diagnostic score for HER2 protein expression (0, 1+, 2+ or 3+) in agreement with the ASCO/CAP guidelines. The automated reading was compared to manual reading of HER2 protein expression and for the borderline (2+) protein expression samples to manual reading of the HER2 gene expression on ﬂuorescence in situ hybridization (FISH). Results: Manual reading demonstrated a sensitivity of 95.5% and a speciﬁcity of 81.0% with 17.0% inconclusive samples. Using the digital HER2-CONNECT test, both sensitivity and speciﬁcity increased (97.7% and 98.1%, respectively), and only 1.9% of the cases were deemed inconclusive. Conclusion: Using the HER2-CONNECT digital image analysis algorithm based on immunohistochemical detection of HER2 protein expression on tumour cell membranes, less than 2% of the cases were inconclusive whether overexpressed
or not. Application of automated image analysis for HER2 protein expression instead of manual reading thus decreases the need for supplementary FISH testing by almost 90%. The impact on cost reduction and turn-around-time is obvious.
O1-4 The inﬂuence of preoperative chemo-radiotherapy on lymph node yield, N-status and survival in rectal cancer results from a nationwide cohort study Jakob Lykke1, Per Jess2, Ole Roikjaer2, On behalf of the Danish Colorectal Cancer Group 1 Department of Surgery, Slagelse Hospital, University of Copenhagen, Faelledvej 1, 4200 Slagelse, Denmark; 2 Department of Surgery, Roskilde Hospital, University of Copenhagen, Koegevej 7-13, Roskilde, Denmark Introduction: In Denmark aaa rectal cancer (RC) patients with T3 tumours with a circumferential resection margin