Mohammed Abu-Hilal

Natural killer cell maturation markers in the human liver and expansion of an NKG2C+KIR+ population

BACKGROUNDnSelected populations of murine natural killer (NK) cells possess memory features to haptens, cytokines, and viruses. Liver-specific adhesion molecules CXCR6 and CD49a have been identified as surface markers in mice. In people, expansion of long-lived terminally differentiated NKG2C+ populations occur in the blood after viral infection. We aimed to compare intrahepatic and blood NK cell receptor expression to determine the existence of CD49a+ and CXCR6+ NK cells in human liver and define the maturation status of NKG2C+ NK cells at this site.nnnMETHODSnTissue samples were taken from the liver margin of 39 patients with hepatic metastases and flushed with chelating buffer followed by collagenase or mechanical digestion. Paired peripheral blood samples were taken from 15 patients, the remainder being unpaired. Mononuclear cells were isolated by ficoll separation and cell surface staining performed for CD3, CD56, CD16, CD57, CD117, CD161, CD158a, CD158b, CD49a, CD49b, CXCR6, NKG2C, and NKp46. Statistical analysis to compare intrahepatic and blood NK cell receptor expression included the median, IQR, and Mann-Whitney U test.nnnFINDINGSnFrequencies of NK cell precursors were similar in the liver and the blood (0·91% [0·62-3·26] vs 0·87 [0·41-1·52]); however, expression of all later markers of maturity were reduced including CD16 (47% [40·4-61·4] vs 88·7 [82·2-93·2], p<0·0001), CD57 (30·7% [25·0-53·9] vs 73·4 [70·4-87·6], p=0·0003), and KIR (11·2% [7·5-14·5] vs 26·7 [17·3-30·8], p<0·0001). Expanded hepatic CD16- NK cells were particularly immature with reduced CD57 and increased CD161 compared with the blood. NKG2C+ NK cells were found in similar frequencies in liver and blood. The hepatic NKG2C+ population was terminally differentiated, as in the circulation, but demonstrated a three-fold increase in KIR expression compared with NKG2C- counterparts, which was not seen in the blood. As in previously published research in mice, CD49a+ and CXCR6+ NK cells were liver resident (6·5% [3·9-14·6] liver vs 2·1 [1·3-4·3] blood, p<0·0001, and 65·3 [48·1-75·2] vs 4·5 [1·43-12·12], p=0·0039, respectively). Both populations were immature, with reduced KIR expression.nnnINTERPRETATIONnWe have shown that the liver contains an expanded population of immature CD16- NK cells. These cells might traffic from the blood and then differentiate into hepatic-specific CD49a+ and CXCR6+ NK cells. The function of these subsets is unknown but their immaturity hints against memory. Terminally differentiated NKG2C+ cells show KIR expansion in the human liver and probably represent an antigen-experienced population, raising the question of whether the liver is a site of NK cell memory acquisition.nnnFUNDINGnMRC Clinical Research Fellowship.

Resolution of paraneoplastic PM/Scl-positive systemic sclerosis after curative resection of a pancreatic tumour

SIR, Compelling evidence has recently been presented that SSc may occur as a paraneoplastic disease, especially for cases expressing hallmark anti-RNA polymerase III autoantibodies [1]. Elegant studies have confirmed expression of variant protein by tumours [2] and larger cohort analyses confirm that the association with malignancy is often contemporaneous with autoantibody development [3]. It would be predicted from this model that tumour removal at an early stage could interrupt the autoimmune process and be associated with resolution of the associated SSc. Such clinical improvement would confirm the role of antigen-driven adaptive autoimmunity driving the disease and strongly support current immunosuppressive treatment strategies. Here we describe the case of a 43-year-old woman with malignancy and clinical features of SSc and polymyositis [creatine kinase (CK)>2000 IU/L] associated with PM/Scl antibodies. The presence of tendon contractures, palmar skin thickening and diffuse distribution of skin disease [peak modified Rodnan skin score (mRSS) 11/51] with generalized hyperpigmentation prompted further radiological assessment for underlying malignancy and a cystic pancreatic lesion with a solid component was identified. A distal pancreatectomy and splenectomy successfully excised all tumour tissue and a diagnosis of solid pseudopapillary pancreatic neoplasm was confirmed on histological and immunohistochemical analysis. Following surgical recovery, the patient was rapidly weaned off all immunosuppression and has remained in clinical remission with resolution of both skin sclerosis (mRSS 4/51) and inflammatory muscle disease [normal CK and Medical Research Council (MRC) grade]. Immunostaining with anti-EXOSC10 antibody (anti-PM/Scl-100) demonstrated increased staining in normal exocrine pancreatic cytoplasm (negative in endocrine pancreas) when compared with tumour from this patient. However, nuclear staining was present universally in tumour tissue and found in very few nuclei in the normal pancreas (Fig. 1, arrows). Clinical remission following full resection and increased nuclear staining for PM/Scl-100 in tumour tissue lends support to a potential pathogenic link in this case, and this has not been previously demonstrated in SSc associated with this antibody subset. To confirm and extend the potential association of this hallmark scleroderma ANA pattern with malignancy we interrogated our SSc research database of 2200 patients and identified 80 patients who tested positive for PM/Scl by Hep-2 immunofluorescence and confirmatory counterimmunoelectrophoresis. Data were available for 70 of these patients, 80% of whom were female, with a mean age of 58.4 years (S.D. 14.0) and a mean age at SSc onset of 44.1 years ( S.D. 14.5). Forty-seven patients (67.1%) had limited cutaneous involvement and the remainder had diffuse disease, except three patients for whom these data were missing. More than one-third of the population showed the presence of calcinosis (38.6%) and inflammatory arthropathy (38.6%), while more than half of the patients in the study population were affected by lung involvement, gastrointestinal involvement and inflammatory myopathy (57.1, 62.9 and 61.4%, respectively). These demographic and clinical data are broadly representative of those published previously for this antibody subset [4]. However, a history of malignancy was found in 14/70 patients (20.0%), with cancer onset within 36 months from SSc diagnosis in 5 patients. This is far FIG. 1 Increased nuclear expression of EXOSC10 (PM/ Scl-100) in tumour tissue of a patient with PM/Scl-positive paraneoplastic scleroderma

IL-12 and IL-15 induce the expression of CXCR6 and CD49a on peripheral natural killer cells: Tissue-homing natural killer cells

Murine hepatic NK cells exhibit adaptive features, with liver‐specific adhesion molecules CXCR6 and CD49a acting as surface markers.

Surgeons' assessment versus risk models for predicting complications of hepato-pancreato-biliary surgery (HPB-RISC): a multicenter prospective cohort study

BACKGROUNDnSeveral studies advise the use of risk models when counseling patients for hepato-pancreato-biliary (HPB) surgery, but studies comparing these models to the surgeons assessment are lacking. The aim of this study was to assess whether risk prediction models outperform surgeons assessment for the risk of complications in HPB surgery.nnnMETHODSnThis prospective study included adult patients scheduled for HPB surgery in three centers in the UK and the Netherlands. Primary outcome was the rate of postoperative major complications. Surgeons assessed the risk prior to surgery while blinded for the formal risk scores. Risk prediction models were retrieved via a systematic review and risk scores were calculated. For each model, discrimination and calibration were evaluated.nnnRESULTSnOverall, 349 patients were included. The rate of major complications was 27% and in-hospital mortality 3%. Surgeons assessment resulted in an AUC of 0.64; 0.71 for liver and 0.56 for pancreas surgery (Pxa0=xa00.020). The AUCs for nine existing risk prediction models ranged between 0.57 and 0.73 for liver surgery and between 0.51 and 0.57 for pancreas surgery.nnnCONCLUSIONnIn HPB surgery, existing risk prediction models do not outperform surgeons assessment. Surgeons assessment outperforms most risk prediction models for liver surgery although both have a poor predictive performance for pancreas surgery.nnnREGISTRATION INFORMATIONnREC reference number (13/SC/0135); IRAS ID (119370). TRIALREGISTER.NL: NTR4649.