SEARCH · PubMed Health

Results for “Prognostic model”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 433 records · Page 24Linked to original sources

PubMed1989

Model predicting survival in stage I melanoma based on tumor progression.

We used the lesional steps in tumor progression and multivariable logistic regression to develop a prognostic model for primary, clinical stage I cutaneous melanoma. This model is 89% accurate in predicting survival. Using histologic criteria, we assigned melanomas to tumor progression steps by ascertaining their particular growth phase. These phases were the in situ and invasive radial growth phase and the vertical growth phase (the focal formation of a dermal tumor nodule or dermal tumor plaque within the radial growth phase or such dermal growth without an evident radial growth phase). After a minimum follow-up of 100.6 months and a median follow-up of 150.2 months, 122 invasive radial-growth-phase tumors were found to be without metastases. Eight-year survival among the 264 patients whose tumors had entered the vertical growth phase was 71.2%. Survival prediction in these patients was enhanced by the use of a multivariable logistic regression model. Twenty-three attributes were tested for entry into this model. Six had independently predictive prognostic information: (a) mitotic rate per square millimeter, (b) tumor-infiltrating lymphocytes, (c) tumor thickness, (d) anatomic site of primary melanoma, (e) sex of the patient, and (f) histologic regression. When mitotic rate per square millimeter, tumor-infiltrating lymphocytes, primary site, sex, and histologic regression are added to a logistic regression model containing tumor thickness alone, they are independent predictors of 8-year survival (P less than .0005).

Female↗

PubMed1996

A risk model for ovarian carcinoma patients using CA 125: time to normalization renders second-look laparotomy redundant.

BACKGROUND: We evaluated the incorporation of CA 125 normalization times into a prognostic model based on pretreatment variables in patients with ovarian carcinoma to determine if they could render second-look laparotomy (SLL) redundant. METHODS: A total of 54 consecutive patients with ovarian carcinoma who underwent SLL between 1985 and 1990 were included in this analysis. At diagnosis, all of the patients had abnormal CA-125 serum levels, which fell to within the normal range during chemotherapy. Cox's model was used to select pretreatment variables relevant for prognosis. The influence of the time to normalization of CA 125 (< or = vs. > 1 months) and the capability of SLL results to modify prognostic prediction, were also evaluated. RESULTS: The size of the residual tumor at the beginning of therapy, and Eastern Cooperative Oncology Group (ECOG) performance status (PS) were independently predictive of survival. The time to normalization of CA 125 serum levels (analyzed either as -a continuous or as a two-category variable) also had an independent prognostic role when included in the model. When we examined the inclusion of both CA 125 parameter and SLL into the model together, we found that only CA 125 continued to have an independent prognostic relevance. On the basis of the two pretreatment parameters (PS and tumor size) and of this response parameter (time to normalization of CA 125 values) we selected six subgroups of patients having different outcomes (log rank test of equality over-strata < 0.001). Patients with good prognostic pretreatment variables, and those with intermediate prognosis at the beginning of therapy who showed a quick normalization of CA 125, had an 80% 5-year survival, compared with 16% 5-year survival in the remaining patients. (P < 0.0001). CONCLUSIONS: Our data suggest that the survival of patients with advanced ovarian carcinoma could be accurately predicted by considering some pretreatment variables and time to CA 125 normalization together, without performing SLL. Our risk model, however, needs to be validated by larger prospective trials, to draw any definitive conclusions about the abandonment of surgically defined response.

Biomarkers, Tumor↗

PubMed2002

Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning.

Diffuse large B-cell lymphoma (DLBCL), the most common lymphoid malignancy in adults, is curable in less than 50% of patients. Prognostic models based on pre-treatment characteristics, such as the International Prognostic Index (IPI), are currently used to predict outcome in DLBCL. However, clinical outcome models identify neither the molecular basis of clinical heterogeneity, nor specific therapeutic targets. We analyzed the expression of 6,817 genes in diagnostic tumor specimens from DLBCL patients who received cyclophosphamide, adriamycin, vincristine and prednisone (CHOP)-based chemotherapy, and applied a supervised learning prediction method to identify cured versus fatal or refractory disease. The algorithm classified two categories of patients with very different five-year overall survival rates (70% versus 12%). The model also effectively delineated patients within specific IPI risk categories who were likely to be cured or to die of their disease. Genes implicated in DLBCL outcome included some that regulate responses to B-cell-receptor signaling, critical serine/threonine phosphorylation pathways and apoptosis. Our data indicate that supervised learning classification techniques can predict outcome in DLBCL and identify rational targets for intervention.

Antineoplastic Combined Chemotherapy Protocols↗

PubMed2005

A model for predicting outcomes in patients with differentiated thyroid cancer and model performance in comparison with other classification systems.

BACKGROUND: This study was performed to determine the prognostic factors for differentiated thyroid cancer, and to establish a mathematical prognostic model. STUDY DESIGN: A retrospective study was conducted in 347 differentiated thyroid cancer patients. Univariate and multivariate prognostic factor analyses were carried out using the Kaplan-Meier and Cox regression methods. RESULTS: Without adjustment for treatment in the multivariate analysis, age, tumor size, angioinvasion, and distant metastasis were significant predictors of outcomes. The very low-risk, low-risk, high-risk, and very high-risk groups were identified from the logistic regression equation. Overall and event-free survival estimations at 10 years were 100% and 100% for very low-risk patients, 88% and 75% for low-risk patients, 30% and 16% for high-risk patients, and 5% and 0% for very high-risk patients. Inclusion of treatment in the multivariate analysis showed, in addition to other variables, that both total or near total thyroidectomy (versus thyroidectomy less than total and near total thyroidectomy, p = 0.0002; hazard ratio, 0.4; 95% CI, 0.3-0.7) and adjuvant radioactive iodine treatment (versus no treatment with radioactive iodine, p = 0.0001; hazard ratio, 0.5; 95% CI, 0.2-0.8) were associated with a reduced hazard of death in the followup period. By subgroup analysis, total and near total thyroidectomy, along with radioactive iodine, appeared to provide a survival benefit for all patients except those in the very low-risk group. CONCLUSIONS: The proposed mathematical model is satisfactory for predicting outcomes. Total and near total thyroidectomy along with radioactive iodine treatment might provide a survival advantage for differentiated thyroid cancer, except for those with very low risk.

Adenocarcinoma, Follicular↗

PubMed2026

Machine learning-guided risk stratification in elderly AML based on genomic, immunophenotypic and therapeutic profiles.

BACKGROUND: Elderly patients with acute myeloid leukemia (AML) exhibit considerable biological and clinical heterogeneity, hindering precise prognosis. Existing prognostic systems inadequately capture the complexity of elderly AML due to their reliance on data from younger cohorts and omission of key factors like immunophenotypic markers and therapeutic profiles. This study aimed to develop and internally validate a machine learning-based prognostic model specifically tailored to elderly AML patients. METHODS: A total of 156 patients were analyzed using a two-stage modeling strategy. Clinical and genomic variables were modeled first, followed by independent analysis of immunophenotypic features. Feature selection was performed using multilayer perceptron (MLP) and random forest (RF), while multivariate Cox regression was used for final model construction. Internal validation was conducted using 1000 bootstrap iterations to assess model stability and performance. RESULTS: The model demonstrated strong predictive performance, with a concordance index (C-index) of 0.702. Time-dependent area under the curve (AUC) and calibration plots confirmed accurate prediction of 1-, 3-, and 5-year overall survival. Decision curve analysis indicated favorable net benefit across a range of threshold probabilities. Key independent prognostic factors identified included TP53 mutations, high CD13 expression, and IDH2 mutations. CONCLUSION: This model provides a robust and interpretable tool for individualized risk stratification in elderly AML. By integrating genomic, immunophenotypic, and therapeutic variables, it may help optimize treatment decisions and improve outcomes for this vulnerable population. Future efforts should focus on external validation and integration of dynamic biomarkers.

Humans↗

PubMed1997

A simple index to predict prognosis independent of axillary node information in breast cancer.

BACKGROUND: Since the course of breast cancer is often unpredictable, we wished to develop a model using characteristics of the primary tumour alone to predict prognosis. METHODS: Several tumour features were determined, and after a median follow-up duration of 65 months, multivariate analysis identified tumour size and grade, oestrogen receptor concentration, axillary lymph node metastasis and tumour cell proliferation fraction (MIB-1 count) as being independently associated with increases in risk for both relapse and death from breast cancer. A prognostic model was constructed using tumour size and grade, oestrogen receptor concentration and MIB-1 count only. A score of 1 for each was given to tumour size > 20 mm, tumour grade 2 or 3, oestrogen receptor concentration < 10 fmol/mg cytosol protein and MIB-1 count > 9%. Five groups established by assigning a combined score of 0, 1, 2, 3 or 4 for each patient were analysed for their associations with disease-free and overall survivals. RESULTS: This preliminary model predicted 5-year survival rates of 97, 91, 85, 68 and 50% for the five groups. The model was further simplified by excluding tumour grade from the analysis. The revised model identified four risk groups with predicted 5-year survival rates of 91, 86, 66 and 52%. This model, the Adelaide prognostic index, was also able to identify four risk groups in both node-negative and node-positive patients. CONCLUSIONS: The Adelaide prognostic index can be used to predict prognosis even in the absence of axillary lymph node information.

Antigens, Nuclear↗

PubMedDate not supplied

Regression modelling strategies for improved prognostic prediction.

Regression models such as the Cox proportional hazards model have had increasing use in modelling and estimating the prognosis of patients with a variety of diseases. Many applications involve a large number of variables to be modelled using a relatively small patient sample. Problems of overfitting and of identifying important covariates are exacerbated in analysing prognosis because the accuracy of a model is more a function of the number of events than of the sample size. We used a general index of predictive discrimination to measure the ability of a model developed on training samples of varying sizes to predict survival in an independent test sample of patients suspected of having coronary artery disease. We compared three methods of model fitting: (1) standard 'step-up' variable selection, (2) incomplete principal components regression, and (3) Cox model regression after developing clinical indices from variable clusters. We found regression using principal components to offer superior predictions in the test sample, whereas regression using indices offers easily interpretable models nearly as good as the principal components models. Standard variable selection has a number of deficiencies.

Coronary Disease↗

PubMed2000

Liver transplantation in primary biliary cirrhosis: risk assessment and 11-year follow-up.

BACKGROUND/AIMS: Liver transplantation (LTx) is the only established treatment in patients with end-stage primary biliary cirrhosis (PBC). Although short-term survival after LTx in this group of patients is usually good, few data exist on the long-term survival. The optimal timing of transplantation is difficult. Thus, the aims of this study were to assess the long-term survival of patients with PBC after LTx and to identify potential predictive factors for a positive outcome. METHODS: Survival of 28 patients with PBC who underwent LTx between 1985 and July 1999 in a single center was studied by Kaplan-Meier analysis and was compared to predicted survival without LTx using established prognostic models for PBC, the Mayo and European risk scores. Potential prognostic parameters obtained before LTx were tested for correlation to survival. Rates of bone fractures as markers of hepatic osteodystrophy were compared before and after LTx. RESULTS: Median follow-up after LTx was 90 months with a maximum of 140 months. Actuarial survival of patients with PBC was 89% after 1, 5, and 10 years and was significantly better than estimated survival without LTx after 1-7 years as calculated by the Mayo and European risk scores. Of several parameters tested, only serum bilirubin and the prognostic scores, but no other liver function tests obtained immediately prior to transplantation were significantly correlated with survival after LTx. The duration of intensive care after LTx was not associated with any parameters obtained before LTx. Bone fractures were diagnosed in 43% of patients of whom the vast majority were osteopenic before LTx as determined by osteodensitometry. CONCLUSION: Long-term survival of a well-defined group of patients with PBC was excellent after LTx and was inversely correlated with preoperative serum bilirubin levels as well as Mayo and European risk scores.

Adult↗

PubMed1997

The Child-Pugh classification as a prognostic indicator for survival in primary sclerosing cholangitis.

Clinical decisions in primary sclerosing cholangitis (PSC) depend upon understanding its variable natural history. Several prognostic models for survival have been developed. We explored the Child-Pugh Classification (CPC) to determine if it predicts survival as well as a disease-specific model (DSM). We identified 208 PSC patients who satisfied established criteria. Prognostic variables were measured from the first visit. Kaplan-Meier survival estimates were obtained for CPC twice, once censoring observations at orthotopic liver transplantation and again considering orthotopic liver transplantation as a failure. A stepwise Cox proportional hazards-model was used to identify the factors associated with mortality. For 208 patients, the mean follow-up was 70 months (SD + 55). Kaplan-Meier 7-year survivals for CPC A, B, and C were 89.8%, 68%, and 24.9%, respectively (P < .001). The Cox model identified CPC and age as the most significant predictors of mortality. Adding the DSM risk score did not significantly improve the fit of the model (P = .75). The following were concluded: 1) CPC is a powerful predictor of survival in PSC; 2) DSM does not enhance the predictive ability of CPC; and 3) CPC may be used as an alternative to the DSM in research studies and clinical decision-making.

Adult↗

PubMed2026

The role of artificial intelligence in the diagnosis and prognosis of traumatic brain injury based on brain CT scans: a systematic review.

Traumatic brain injury (TBI) is a leading cause of emergency department visits and a major contributor to injury-related mortality and long-term neurological disability. Non-contrast computed tomography (CT) is the gold-standard imaging modality for the rapid diagnosis of TBI. Clinical outcomes depend strongly on early detection and prompt acute management. Artificial intelligence (AI)-based models may support faster automated identification of traumatic findings and early prediction of patient prognosis. A systematic literature search was conducted in PubMed/MEDLINE, Scopus, IEEE Xplore, ACM Digital Library, and the Cochrane Library in accordance with PRISMA 2020 guidelines to evaluate AI-based models for automated detection of TBI-related findings on CT and for prediction of clinical outcomes. Risk of bias and applicability were assessed using QUADAS-2 for diagnostic accuracy studies and PROBAST + AI for prediction model studies. Twenty-two studies were included. Sixteen studies evaluated diagnostic tasks and 10 evaluated prognostic outcomes, with four studies contributing to both categories. Diagnostic performance was generally high, with many studies reporting AUC values approaching or exceeding 0.90, particularly for larger lesion volumes.Prognostic performance was more variable, with moderate to high discrimination and substantial heterogeneity. Only 9 studies incorporated independent external validation, and performance was frequently lower in external cohorts. All prognostic model studies were judged to be at high overall risk of bias using PROBAST + AI, and most diagnostic accuracy studies also demonstrated high or unclear risk of bias in at least one QUADAS-2 domain, most frequently in patient selection. AI-based models applied to brain CT demonstrate strong technical performance for both diagnostic and prognostic tasks in TBI. However, most studies relied on retrospective designs and lacked independent external validation which limits models generalizability and raises concern for potential overfitting. Prospective, multicenter studies with standardized methodologies and rigorous external validation are required before widespread clinical implementation.

Humans↗

PubMed2005

Improvement of breast cancer relapse prediction in high risk intervals using artificial neural networks.

The objective of this study is to compare the predictive accuracy of a neural network (NN) model versus the standard Cox proportional hazard model. Data about the 3811 patients included in this study were collected within the 'El Alamo' Project, the largest dataset on breast cancer (BC) in Spain. The best prognostic model generated by the NN contains as covariates age, tumour size, lymph node status, tumour grade and type of treatment. These same variables were considered as having prognostic significance within the Cox model analysis. Nevertheless, the predictions made by the NN were statistically significant more accurate than those from the Cox model (p < 0.0001). Seven different time intervals were also analyzed to find that the NN predictions were much more accurate than those from the Cox model in particular in the early intervals between 1-10 and 11-20 months, and in the later one considered from 61 months to maximum follow-up time (MFT). Interestingly, these intervals contain regions of high relapse risk that have been observed in different studies and that are also present in the analyzed dataset.

Adult↗

PubMed1999

Prediction of benefit from carotid endarterectomy in individual patients: a risk-modelling study. European Carotid Surgery Trialists' Collaborative Group.

BACKGROUND: Carotid endarterectomy lowers the risk of carotid territory ipsilateral ischaemic stroke, and is the treatment of choice, in patients with recently symptomatic 70-99% carotid stenosis. However, the 3-year risk of stroke on medical treatment alone is only about 20%. We investigated whether the efficacy of endarterectomy would be improved if patients with a high risk of stroke on medical treatment and a low risk of operative stroke or death could be identified. METHODS: We developed two prognostic models from data on patients with 0-69% carotid stenosis in the European Carotid Surgery Trial (ECST). The medical model predicted risk of ipsilateral carotid territory major ischaemic stroke (fatal or lasting longer than 7 days) on medical treatment and the surgical model predicted risk of major stroke and death within 30 days of endarterectomy. From these models we developed a prognostic score to identify patients with a high risk of stroke on medical treatment but a low operative risk. We validated the models and tested the scoring system on 990 ECST patients with 70-99% carotid stenosis assigned surgery (594) or medical treatment only (396). FINDINGS: When patients with 70-99% stenosis were stratified by the scoring system, which was based on seven independent prognostic factors, endarterectomy was beneficial in only 162 (16%) patients. The 5-year absolute risk of carotid territory ipsilateral major ischaemic stroke, operative major stroke, or death was lowered by 33% in the 16% of patients with a score of 4 or more (odds ratio 0.12 [95% CI 0.05-0.29], p<0.0001), but not in the other 828 (84%) patients (1.00 [0.65-1.54], p=0.7). INTERPRETATION: Many patients with recently symptomatic 70-99% carotid stenosis may not benefit from carotid endarterectomy. Validation of the predictive score is needed on external datasets, but risk-factor modelling could be useful to identify those patients in whom endarterectomy will be beneficial.

Actuarial Analysis↗

PubMed1999

Predictive model of functional independence in stroke patients admitted to a rehabilitation programme.

OBJECTIVE: To develop a prognostic model to estimate the probability of patients being independent in ambulation and in activities of daily living (ADL) after six months of stroke. DESIGN: Cohort analytical study. SETTING: Rehabilitation departments of two district general hospitals. SUBJECTS: Ninety-two consecutive stroke patients admitted to a rehabilitation programme. MAIN OUTCOME MEASURES: Independent ambulation was defined as a Functional Ambulation Classification (FAC) > or =4, and the independence in ADL as a Barthel Index (BI) > or =285. All patients were assessed on admission to rehabilitation, and in the first, second, third, fourth and six months after stroke. RESULTS: Prognostic factors were identified by means of a multivariate survival analysis using Cox regression. Three variables were predictors for a FAC > or =4: (1) The patients in the motor (M), motor-sensitive (MS) and motor-sensitive with hemianopsia (MSH) groups (relative risk (RR) 5.43 of M with respect to MSH, and 2.41 of MS to MSH). (2) A Motricity Index >25 (RR 3.19). (3) An age <70 years old (RR 1.99). For a BI > or =85 three predictors were selected: (1) The classification M-MS-MSH (RR 6.02 M to MSH, and 1.52 MS to MSH). (2) An initial BI >20 (RR 3.45); the highest contribution in the achievement of an initial BI >20 was bowel and bladder continence. (3) The antecedent of previous independence (RR 2.68). The predictive models, constructed by means of multiple logistic regression correctly classified 77% and 79% of the patients who obtained FAC > or =4 and a BI > or =85 respectively. CONCLUSIONS: The syndromic classification M, MS and MSH, together with other routinely available data, such as the Motricity Index, BI, the age and the previous functionality, can be used to obtain a patient prognosis level with regard to ambulation and ADL independence.

Activities of Daily Living↗

PubMed1988

Stroke incidence and risk factors for stroke in Copenhagen, Denmark.

Stroke incidence in Copenhagen, Denmark was recorded in a random population sample of 19,327 persons invited for two health examinations with 5 years' interval from 1976 to 1983. Stroke incidence increased exponentially with age. After adjustment to the age and sex distribution of the Danish population in 1980, the estimated incidence of first stroke was 1.41/1000 women and 2.48/1000 men; the total incidence was 1.94/1000 population. Risk factor analysis was based on the initial examination of 13,088 persons greater than 35 years old without previous stroke who responded to the first invitation, in whom 295 first strokes were subsequently observed. We used the regression model of Cox. However, our use of this model differs from the somewhat automatic procedures normally used to develop prognostic models. Evaluation of the causative effect of a particular risk factor requires that the direction of mutual influences between the factor in question and other risk factors is established/postulated. Among the 16 potential risk factors for stroke we examined, significant effects were found for age, sex, household income, smoking habits, systolic blood pressure, diabetes, plasma cholesterol concentration, ischemic heart disease, and atrial fibrillation. No significant effect could be demonstrated for a positive family history of stroke, years of school education, marital status, alcohol consumption, daily use of tranquilizers, body mass index, or postmenopausal hormone treatment.

Adult↗

PubMed1997

PROGNOSTICATION OF BREAST CANCER BY MULTIVARIATE METHODS.

Three multivariate prognostic models based on Cox's regression were tested in terms of how they predicted prognosis in a material of 134 patients with breast cancer. The multivariate models all incorporated tumor size, mitotic activity index (MAI), and axillary lymph node status in their formulas, and were originally produced through studies on different patient materials. The predictive behavior of MAI was also tested separately in the same material. The multivariate models gave roughly parallel predicted percentages of survival at two years (CV=5.2%), but showed clearly greater variation later (12.3% and 24.4% at 5 and 9 years, respectively). The results were more uniform between the multivariate models, than between the prediction by MAI and the multivariate models. The variation between repeated estimates was smaller within multivariate models than within the estimation of one of their components (MAI). We found the use of the multivariate models easy. However, traditional hospital practice does not necessarily favor the use of multivariate models, although they seem to group patients more reliably than single prognostic features.

Journal Article↗

PubMed2004

Prediction of an ongoing pregnancy after intrauterine insemination.

OBJECTIVE: To develop a prognostic model for the outcome of IUI. DESIGN: Retrospective cohort study. SETTING: Four fertility centers in The Netherlands. PATIENT(S): Couples of whom the female partners had a regular cycle and who had been treated with IUI. INTERVENTION(S): Intrauterine insemination with and without ovarian hyperstimulation. MAIN OUTCOME MEASURE(S): Ongoing pregnancy. RESULT(S): Overall, 3371 couples were included who underwent 14968 cycles. There were 1229 (8.2%) pregnancies, of which 1000 (6.7%) pregnancies were ongoing. Logistic regression analysis demonstrated that increasing maternal age, longer duration of subfertility, presence of male factor subfertility, one-sided tubal pathology, endometriosis, uterine anomalies, and an increasing number of cycles were unfavorable predictors for an ongoing pregnancy. Cervical factor and the use of ovarian hyperstimulation were favorable predictors. The area under the receiver operating characteristic curve was 0.59. When couples were divided into four categories based on prognosis, the difference between the predicted and observed chance, that is, the calibration, was less than 0.5% in each of the four groups. CONCLUSION(S): Although our model had a relatively poor discriminative capacity, data on calibration showed that the selected prognostic factors allow distinction between couples with a poor prognosis and couples with a good prognosis. After external validation, this model could be of use in patient counseling and clinical decision making.

Adult↗

PubMed2006

Prediction of risk of seizure recurrence after a single seizure and early epilepsy: further results from the MESS trial.

BACKGROUND: The MRC Multicentre trial for Early Epilepsy and Single Seizures (MESS) showed a reduced risk of further seizures in patients, for whom treatment with antiepileptic drugs was uncertain, who were randomly assigned immediate treatment compared with delayed treatment. However, there was no evidence of an effect on [corrected] long-term remission rates. This study was undertaken to assess the role of patient characteristics and treatment in the prediction of seizure recurrence. This will enable decision-making on the basis of the perceived risk of treatment compared with the benefit of reducing the risk of further seizures in the initial years after diagnosis. METHODS: A prognostic model was developed based on individual patient data from MESS to enable identification of patients at low, medium, or high risk of seizure recurrence. A split-sample approach was used in which the model was developed on a subsample of the full data and validated on the remainder of the sample. Distinction of the prognostic groups and predictive accuracy of the model were assessed. FINDINGS: Number of seizures of all types at presentation, presence of a neurological disorder, and an abnormal electroencephalogram (EEG) were significant factors in indicating future seizures. Individuals with two or three seizures, a neurological disorder, or an abnormal EEG were identified as the medium-risk group, those with two of these features or more than three seizures as the high-risk group, and those with a single seizure only as the low-risk group. INTERPRETATION: The model shows that there is little benefit to immediate treatment in patients at low risk of seizure recurrence, but potentially worthwhile benefits are seen in those at medium and high risk.

Adolescent↗

PubMed2001

The HLA-DR3,DQ2 heterozygous genotype is associated with an accelerated progression of primary sclerosing cholangitis.

BACKGROUND: An improvement of prognostic models in primary sclerosing cholangitis (PSC) is needed. In particular, inclusion of prognostic markers that are independent of the disease stage would be advantageous. We investigated whether HLA class II genes associated with PSC are also related to disease progression. METHODS: The study included 265 PSC patients from five European countries with a median follow-up of 9.1 years. The end-points were death (n = 38) or liver transplantation (n = 52). Thirty patients developed cholangiocarcinoma during follow-up. RESULTS: The DRB1*03,DQA1*0501, DQB1*02 (i.e. DR3,DQ2) heterozygous genotype was associated with an increased risk of death or liver transplantation (hazard ratio = 1.63; 95% confidence interval (CI) = 1.06-2.52). The presence of a DQ6 encoding haplotype (DQB1*0603 or DQB1*0602) in DR3,DQ2 negative individuals was associated with a reduced risk of death or liver transplantation (hazard ratio = 0.57; 95% CI = 0.36-0.88). There was a trend towards an increased risk of developing cholangiocarcinoma among DR4,DQ8 positive patients, but this did not reach significance (odds ratio = 2.27; 95% CI = 0.78-6.62). CONCLUSION: The DR3,DQ2 heterozygous genotype is associated with a more rapid progression of PSC, whereas HLA-DQ6 is associated with a retarded disease progression. It is possible that the DR4,DQ8 haplotype is related to cholangiocarcinoma development.

Adolescent↗