An aetiological role of human papillomavirus (HPV) and/or human polyomaviruses (HPyVs) has been proposed in adenoid cystic carcinoma (AdCC). Moreover, HPV-related multiphenotypic carcinoma (HMSC) was recently introduced as an emerging entity of the sinonasal region. Here, we primarily want to study the role of HPV/HPyV in a large AdCC cohort and, secondly, possibly identify and characterize HMSC. Tumour DNA from 68 patients initially diagnosed with AdCC between 2000 and 2012 was, therefore, tested for 27 HPV types and 10 HPyVs. HPV DNA-positive samples were micromorphologically re-evaluated, further stained for p16INK4a, S100, p63 and CD117 and tested for the presence of the MYB-NFIB fusion transcript. Notably, no samples were HPyV-positive, while one sinonasal and two tonsillar carcinomas were HPV- and p16-positive. After re-evaluating the micromorphology, immunohistochemistry and presence of fusion transcripts, all tumours had the same appearance and fitted within the diagnosis of HMSC, but in all these three cases, the morphology of the HMSC and basaloid squamous cell carcinoma was overlapping. We conclude that HPV and HPyV have no major role in AdCC. However, based on our data, we also suggest that HMSC should be considered as a basaloid variant of squamous cell carcinoma, and not its own entity, until better characterized.

BackgroundAutoantibodies to citrullinated protein (ACPA; measured as anti-CCP; aCCP) and rheumatoid factor (RF) appear years before clinical onset of RA and are essential tools in today’s classification criteria for RA. In animal models, antibodies to joint specific proteins (JP) can induce arthritis, and they are also present at onset of RA [1]. As there is a need for increased precision for early diagnosis of RA as well as identification of different subtypes of the disease, we aim to assess whether autoantibodies to native or modified JP can be used for early and precise diagnosis of RA.ObjectivesTo study whether antibodies to JP, alone or in combination with ACPA/RF, could increase the diagnostic sensitivity and specificity in untreated early (ue)RA patients.MethodsAntibodies to JP were analysed in serum from patients in three independent ueRA cohorts as well as from population controls without rheumatic diseases (WINGA, Gothenburg and MFM-ÅUS, Malmö n=1062). ERAp (n=66), the smallest and most recent cohort was chosen for screening, and BARFOT and TIRA-2 (n=1939) for validation. We have developed a bead-based multianalyte flow immunoassay [2] and screened approx. 350 peptides derived from JPs of interest. We included monoclonal antibodies as assay calibrators and determined limit of detection (LoD). To assess positivity for autoantibodies to JP of interest above LoD, we used 5MAD (median absolute deviation) of the control populations as the cut-off.ResultsIn the ERAp cohort, 5 autoantibodies discriminated RA patients from controls with 81% sensitivity and 100% specificity (Table 1). The same autoantibodies had 68% sensitivity and 98% specificity in the combined BARFOT and TIRA-2 cohorts. Together with RF and aCCP, only 2 of the 5 autoantibodies added statistically significant diagnostic value, increasing the sensitivity from 48% to 61% with 99% specificity. In aCCP- and RF-negative ueRA patients (n=536), the novel biomarkers identified 22.5% of the patients with 99% specificity compared to controls.Table 1.Diagnostic capacity of the joint-specific antibodiesTest panelPerformanceGroup of patientsaCCP+RF+JP+SensitivitySpecificityAUC(ROC)ERApAll patients (n=66)--X81%100%89%RF and aCCP-neg patients (n=7)1------BARFOT and TIRA-2, combined dataAll patients (N=1939)--X68%98%86%All patients (N=1939)X--58%99%78%All patients (N=1939)2XX-48%100%84%All patients (N=1939)2, 3XXX61%99%86%RF and/or aCCP-pos patients (N=1403)--X84%99%93%RF and aCCP-neg patients (N=536)--X22%99%67%RA, literature valuesAnti-CCP testXN/AN/A53–71%95–96%N/A1Not analysed due to lack of power2This patient population is both aCCP+ and RF+3Only 2 of the 5 autoantibodies added statistically significant to the diagnostic valueAUC, Area under the curve; ROC, receiver operating characteristic curve; N/A, not applicable. Controls without rheumatic diseases: N=935 for BARFOT / TIRA-2 and N=27 for ERAp.ConclusionAutoantibodies to JP discriminate ueRA patients better then aCCP and RF alone and add an increased diagnostic value in particular for seronegative patients.References[1]Holmdahl, R., V. Malmstrom, and H. Burkhardt, Autoimmune priming, tissue attack and chronic inflammation - the three stages of rheumatoid arthritis. Eur J Immunol, 2014. 44(6): p. 1593-9.[2]Viljanen, J., et al., Synthesis of an Array of Triple-Helical Peptides from Type II Collagen for Multiplex Analysis of Autoantibodies in Rheumatoid Arthritis. ACS Chem Biol, 2020. 15(9): p. 2605-2615. Correction: ACS Chem Biol, 2020. 15(11): p. 3072AcknowledgementsBARFOT study group.Disclosure of InterestsErik Lönnblom: None declared, Monica Leu Agelii: None declared, Outi Sareila Employee of: Part time employee in Vacara AB, Ingiäld Hafström: None declared, Maria Andersson: None declared, Lei Cheng: None declared, Göran Bergström: None declared, Anna-Karin H Ekwall: None declared, Anna Rudin: None declared, Alf Kastbom: None declared, Christopher Sjowall: None declared, Bingze Xu: None declared, Lennart T.H. Jacobsson: None declared, Johan Viljanen: None declared, Jan Kihlberg: None declared, Inger Gjertsson: None declared, Rikard Holmdahl Shareholder of: Rikard Holmdahl the founder of Vacara AB.