Surveillance pouchoscopy for dysplasia: Cleveland Clinic Ileoanal Pouch Anastomosis Database

BACKGROUND

No formal guidelines exist for surveillance pouchoscopy following ileal pouch-anal anastomosis (IPAA) for ulcerative colitis.

METHODS

All adults who had previously had IPAA for ulcerative colitis, and underwent a pouchoscopy between 1 January 2010 and 1 January 2020, were included.

RESULTS

A total of 9398 pouchoscopy procedures were performed in 3672 patients. The majority of the examinations were diagnostic (8082, 86·0 per cent; 3260 patients) and the remainder were for routine surveillance (1316, 14·0 per cent; 412 patients). Thirteen patients (0·14 per cent of procedures) were found to have biopsy-proven neoplasia at the time of pouchoscopy; seven had low-grade dysplasia (LGD) (0·07 per cent; all located in the anal transition zone), none had high-grade dysplasia (HGD) and six (0·06 per cent) had invasive adenocarcinoma (4 in anal transition zone and 6 in pouch). Of the six patients with adenocarcinoma, four had neoplasia at the time of proctocolectomy (2 adenocarcinoma, 1 LGD, 1 HGD); all six were symptomatic with anal bleeding or pelvic pain at the time of pouchoscopy, had a negative surveillance pouchoscopy examination within 2 years of diagnosis of adenocarcinoma, had palpable masses on digital rectal examination, and had visible lesions at the time of pouchoscopy.

CONCLUSION

Surveillance pouchoscopy is not recommended in asymptomatic patients because significant neoplasia following IPAA for ulcerative colitis is rare.

Building Clinical Classifiers Using Incomplete Observations – A Neural Network Ensemble for Hepatoma Detection in Patients with Cirrhosis

One objective of liver transplant evaluation is to identify patients that harbor a hepatoma, but standard screening techniques are not sensitive enough. We trained neural network ensembles to predict the presence of hepatoma in patients with cirrhosis, based on information collected at the time of transplant evaluation. Network architecture and training were modified to handle missing observations. Three ensembles were trained: ensemble A using the subset with no missing observations (528 patients); ensemble B using the complete set, which included missing observations (853 patients); and ensemble C using the smaller subset, originally with complete data, but after a fixed number of observations were deleted (i. e., made “missing”). Ensemble performance on testing sets was very good. The areas under the ROC curves were 0.91, 0.89, and 0.90, for ensembles A. B, and C, respectively. Neural networks can successfully perform this classification task, and strategies can be developed that allow use of incomplete observations.

Performance of Multi-Layer Feedforward Neural Networks to Predict Liver Transplantation Outcome

A novel multisolutional clustering and quantization (MCO) algorithm has been developed that provides a flexible way to preprocess data. It was tested whether it would impact the neural network’s performance favorably and whether the employment of the proposed algorithm would enable neural networks to handle missing data. This was assessed by comparing the performance of neural networks using a well-documented data set to predict outcome following liver transplantation. This new approach to data preprocessing leads to a statistically significant improvement in network performance when compared to simple linear scaling. The obtained results also showed that coding missing data as zeroes in combination with the MCO algorithm, leads to a significant improvement in neural network performance on a data set containing missing values in 59.4% of cases when compared to replacement of missing values with either series means or medians.

Antimicrobial peptides: mediators of innate immunity as templates for the development of novel anti-infective and immune therapeutics

Antimicrobial molecules are ancient and essential small cationic molecules of the host defence system which are found in a wide variety of species. They display antimicrobial activity against a wide range of bacteria, fungi and viruses, an activity that has been mostly attributed to the disruption of microbial membranes. In this article, we will review the "classical" functions of 3 classes of antimicrobial molecules, namely defensins, cathelicidins, and the four-disulfide core proteins secretory leukocyte proteinase inhibitor (SLPI) and elafin. In addition to the study of their expression in a variety of cell types and the regulation of their production, we will also describe novel properties of these molecules that have been highlighted by recent studies. These include their ability to chemoattract a variety of inflammatory, immune and other cell types (neutrophils, macrophages, monocytes, lymphocytes, mast cells, epithelial cells) in vitro and in vivo. In addition, we will discuss the potential use of these newly discovered properties for therapeutic or vaccination purposes, using protein- or gene-transfer based methodologies. Finally, we will examine in an extensive fashion the strategies used by microorganisms to circumvent and subvert host defence mechanisms, such as the modifications of cell membranes and walls, the secretion of inactivating proteins and proteases and the down-regulation of expression of antimicrobial molecules. Increased understanding of the mechanisms used by both the host and the microbes to 'win the battle' may ultimately lead to new therapeutic strategies aimed to treat infectious diseases.

Stimulated response of peripheral lymphocytes may distinguish cyclosporine effect in renal transplant recipients receiving a cyclosporine+rapamycin regimen

BACKGROUND

Clinically, cyclosporine (CSA, Neoral) is titrated to concentrations, and not to pharmacological effect.

METHODS

Intracellular interleukin- (IL) 2 was measured in phorbol myristic acid-ionomycin-stimulated peripheral lymphocytes by flow cytometry, after isolation from 14 renal transplant recipients receiving CSA+prednisone, and double-blind rapamycin (rapamycin:placebo=4:1).

RESULTS

The proportion (%) of CD4+IL-2+ lymphocytes corresponding to CSA levels (mean+/-SD ng/ml) measured preoperatively (TO=O), and on postoperative day 8, before (356+/-63), and 2 hr after the morning dose (Cmax=1567+/-669), decreased from 39+/-16 to 15+/-8 and 3+/-1.6, respectively. Reciprocally, unresponsive lymphocytes (%CD4+IL-2-) increased with increasing CSA levels and predicted an EC50 of 249 ng/ml (CSA concentration at which CD4+IL-2- cells increased by 50% over baseline) in an Emax pharmacodynamic model.

CONCLUSIONS

Clinically, the pharmacological effect of CSA is quantifiable, and lies in the upper end of the predicted range. In our Neoral-treated sample population, Cmax was associated with the least variable "cyclosporine effect." Such information could potentially individualize immunosuppression, and lead to rational dosing strategies.

A prospective randomized trial of tacrolimus and prednisone versus tacrolimus, prednisone, and mycophenolate mofetil in primary adult liver transplant recipients: an interim report

BACKGROUND

Tacrolimus (Tac) and mycophenolate mofetil (MMF) are newly approved immunosuppressive agents. However, the safety and efficacy of the combination of MMF and Tac in primary liver transplantation has not been determined.

METHODS

An Institutional Review Board-approved, open-label prospective randomized protocol was initiated to study the efficacy and toxicity of Tac and steroids (double-drug therapy) versus Tac, steroids, and MMF (triple-drug therapy) in primary adult liver transplant recipients. Both groups of patients began on the same doses of Tac and steroids. Patients randomized to triple-drug therapy also received 1 g of MMF twice a day.

RESULTS

Between August 1995 and January 1997, 200 patients were enrolled, 99 in double-drug therapy and 101 in triple-drug therapy. All patients were followed until May 1997, with a mean follow-up of 12.7 months. During the study period, 28 of 99 patients in double-drug therapy received MMF to control ongoing acute rejection, nephrotoxicity, and/or neurotoxicity. On the other hand, 61 patients in triple-drug therapy discontinued MMF for infection, myelosuppression, and/or gastrointestinal disturbances. By an "intention-to-treat analysis," the actuarial 1-year patient survival rate was 85.1% in double-drug therapy and 83.1% in triple-drug therapy (P=0.77). The actuarial 1-year graft survival rate was 80.2% for double-drug therapy and 79.2% for triple-drug therapy (P=0.77). Forty-one patients (41.4%) in double-drug therapy and 32 (31.7%) in triple-drug therapy had at least one episode of rejection, but this was not statistically significant (P=0.15). The mean maintenance dose of corticosteroids was slightly lower in triple-drug compared with double-drug therapy.

CONCLUSION

Patient and graft survival rates were similar in both groups. There was a trend to a lower incidence of rejection, reduced nephrotoxicity, and a lesser amount of maintenance corticosteroids in triple-drug therapy compared with double-drug therapy.

Clinical outcome of patients infected with hepatitis C virus infection on survival after primary liver transplantation under tacrolimus

The outcome of hepatitis C virus (HCV) infection on patient and graft survival after orthotopic liver transplantation (OLT) has been controversial. An earlier experience with a higher dose of tacrolimus (>/=0.1 mg/kg/d intravenously and >/=0.2 mg/kg/d orally) was associated with a worse clinical outcome in patients infected with HCV. The clinical outcome of 183 liver transplant recipients with end-stage liver disease (ESLD) secondary to HCV infection (HCV group) was compared with a contemporary cohort of 556 patients with HCV infection who underwent transplantation for nonviral, nonmalignant ESLD (control group). All patients were prospectively screened for anti-HCV antibodies and HCV RNA by reverse-transcriptase polymerase chain reaction. All OLT patients were receiving low-dose tacrolimus immunosuppression. Cumulative patient survival rates for the HCV group were 80% after 1 year and 75% after 3 years compared with rates of 84% and 78%, respectively, in the control group (P = .452). Primary graft survival rates at the same time intervals for the HCV group and the control group were 72% and 77.5% at 1 year and 67% and 72% at 3 years, respectively (P = .144). The incidence of re-transplantation (re-OLT) in the HCV group and the control group was 12.6% and 10.4%, respectively (P = .42). Chronic HCV infection as an indication for OLT with a lower dose of tacrolimus immunosuppression (</=0.05 mg/kg/d intravenously and </=0.1 mg/kg/d orally) is associated with a similar patient and graft survival as those without HCV infection.

In vitro effects and clinical evaluation of a human chorionic gonadotrophin preparation in acute leukemia

Commercial human chorionic gonadotrophin (HCG) preparations decrease the tumorigenicity of human tumors in immunodeficient mice and induce apoptotic cell death in animal tumor models. Preliminary studies in humans have demonstrated tumor regression in patients with Kaposi's sarcoma given intralesional injections of HCG. To further evaluate HCG's antitumor activity we conducted in vitro and clinical evaluations of HCG in acute myeloid leukemia (AML). In HL-60 leukemic cell lines, a 20-40% inhibition of cell density was demonstrated by trypan blue exclusion method at low concentrations of an HCG preparation (2 x 10(-3)-2 x 10(-2)). Similar concentrations also resulted in a reduction in the proportion of cells in G2M phase of the cell cycle, as well as enhanced differentiation compared to control cells. Fifteen patients with advanced AML with marrow blast counts >30%, and five with marrow blast counts between 10 and 26% were given daily subcutaneous injections of HCG 2-4 IU and oral levamisole 50 mg weekly. Five patients with absolute blast counts in the blood ranging from 0 to 3500/microl and percent blasts in the marrow ranging from 16 to 81% were observed to have no progressive increase in either marrow or peripheral blast counts for 70-121 days. One patient with a pretreatment blast count of 10% in the marrow, no circulating blasts and minor cytopenias had a decrease in marrow blasts to less than 5% which has persisted at 550 days. No significant improvement from baseline levels of neutrophils, hemoglobin or platelets were observed in any nl the patients treated. Increases in apoptotic cell death were observed in over 50% of patients' cells with some demonstrating peak levels similar to experiences in patients treated with DNA-damaging chemotherapy. A decreased expression of bcl-2 was seen in the majority of patients ranging from 6 to 62%. These new observations suggest that HCG preparations may inhibit leukemic cell growth through enhancement of cell death mechanisms and could be used in judicious combinations with other approaches. The results confirm the pro-apoptotic effects of HCG preparations reported in patients with Kaposi's sarcoma. Identification of the active component of HCG preparations and further understanding of its growth modulatory action will be important in its development as a clinically useful agent.

Clinical intestinal transplantation: new perspectives and immunologic considerations

BACKGROUND

Although tacrolimus-based immunosuppression has made intestinal transplantation feasible, the risk of the requisite chronic high-dose treatment has inhibited the widespread use of these procedures. We have examined our 1990-1997 experience to determine whether immunomodulatory strategies to improve outlook could be added to drug treatment.

STUDY DESIGN

Ninety-eight consecutive patients (59 children, 39 adults) with a panoply of indications received 104 allografts under tacrolimus-based immunosuppression: intestine only (n = 37); liver and intestine (n = 50); or multivisceral (n = 17). Of the last 42 patients, 20 received unmodified adjunct donor bone marrow cells; the other 22 were contemporaneous control patients.

RESULTS

With a mean followup of 32 +/- 26 months (range, 1-86 months), 12 recipients (3 intestine only, 9 composite grafts) are alive with good nutrition beyond the 5-year milestone. Forty-seven (48%) of the total group survive bearing grafts that provide full (91%) or partial (9%) nutrition. Actuarial patient survival at 1 and 5 years (72% and 48%, respectively) was similar with isolated intestinal and composite graft recipients, but the loss rate of grafts from rejection was highest with intestine alone. The best results were in patients between 2 and 18 years of age (68% at 5 years). Adjunct bone marrow did not significantly affect the incidence of graft rejection, B-cell lymphoma, or the rate or severity of graft-versus-host disease.

CONCLUSIONS

These results demonstrate that longterm rehabilitation similar to that with the other kinds of organ allografts is achievable with all three kinds of intestinal transplant procedures, that the morbidity and mortality is still too high for their widespread application, and that the liver is significantly but marginally protective of concomitantly engrafted intestine. Although none of the endpoints were markedly altered by donor leukocyte augmentation (and chimerism) with bone marrow, establishment of the safety of this adjunct procedure opens the way to further immune modulation strategies that can be added to the augmentation protocol.

HLA and cross-reactive antigen group matching for cadaver kidney allocation

BACKGROUND

Allocation of cadaver kidneys by graded human leukocyte antigen (HLA) compatibility scoring arguably has had little effect on overall survival while prejudicing the transplant candidacy of African-American and other hard to match populations. Consequently, matching has been proposed of deduced amino acid residues of the individual HLA molecules shared by cross-reactive antigen groups (CREGs). We have examined the circumstances under which compatibility with either method impacted graft survival.

METHODS

Using Cox proportional hazards regression modeling, we studied the relationship between levels of conventional HLA mismatch and other donor and recipient factors on primary cadaver kidney survival between 1981 and 1995 at the University of Pittsburgh (n=1,780) and in the United Network for Organ Sharing (UNOS) Scientific Registry during 1991-1995 (n=31,291). The results were compared with those obtained by the matching of amino acid residues that identified CREG-compatible cases with as many as four (but not five and six) HLA mismatches.

RESULTS

With more than one HLA mismatch (> 85% of patients in both series), most of the survival advantage of a zero mismatch was lost. None of the HLA loci were "weak." In the UNOS (but not Pittsburgh) category of one-HLA mismatch (n=1334), a subgroup of CREG-matched recipients (35.3%) had better graft survival than the remaining 64.7%, who were CREG-mismatched. There was no advantage of a CREG match in the two- to four-HLA incompatibility tiers. Better graft survival with tacrolimus was observed in both the Pittsburgh and UNOS series.

CONCLUSIONS

Obligatory national sharing of cadaver kidneys is justifiable only for zero-HLA-mismatched kidneys. The potential value of CREG matching observed in the one-HLA-mismatched recipients of the UNOS (but not the Pittsburgh) experience deserves further study.

Tacrolimus rescue therapy for renal allograft rejection--five-year experience

Over the 5 year period from 7/14/1989 until 5/24/1994, we have attempted graft salvage with tacrolimus conversion in a total of 169 patients (median age 33 years, range 2-75 years) with ongoing rejection on baseline CsA immunosuppression after failure of high dose corticosteroids and/or antilymphocyte preparations to reverse rejection. The indications for conversion to tacrolimus were ongoing, biopsy confirmed rejection in all patients. The median interval to tacrolimus conversion was 2 months (range 2 days to 55 months; mean 4.3+/-2.6 months) after transplantation. All patients had failed high dose corticosteroid therapy and 144 (85%) of the 169 patients had received at least one course of an antilymphocyte preparation plus high dose corticosteroid therapy prior to conversion. Twenty-eight patients (17%) were dialysis-dependent at the time of conversion owing to the severity of rejection. With a mean follow-up of 30.0+/-2.4 months (median 36.5 months, range 12-62 months), 125 of 169 patients (74%) have been successfully rescued and still have functioning grafts with a mean serum creatinine (SCR) of 2.3+/-1.1 mg/dl. Of the 144 patients previously treated with antilymphocyte preparations, 117 (81%) were salvaged. Of the 28 patients on dialysis at the time of conversion to tacrolimus, 13 (46%) continue to have functioning grafts (mean SCR 2.15+/-0.37 mg/dl) at a mean follow-up of 37.3+/-16.7 months. In the 125 patients salvaged, prednisone doses have been lowered from 28.0+/-9.0 mg/d (median 32, range 4-60 mg/d) preconversion to 8.5+/-4.1 mg/d (median 12 mg/d, range 2.5-20 mg/d) postconversion. Twenty-eight patients (22.4%) are currently receiving no steroids. This 5 year experience demonstrates that tacrolimus has sustained efficacy as a rescue agent for ongoing renal allograft rejection. Based on these data, we recommend that tacrolimus be used as an alternative to the conventional drugs used for antirejection therapy in renal transplantation.

Weaning of immunosuppression in liver transplant recipients

Immunosuppression has been sporadically discontinued by noncompliant liver allograft recipients for whom an additional 4 1/2 years of follow-up is provided. These anecdotal observations prompted a previously reported prospective drug withdrawal program in 59 liver recipients. This prospective series has been increased to 95 patients whose weaning was begun between June 1992 and March 1996, 8.4+/-4.4 (SD) years after liver replacement. A further 4 1/2 years follow-up was obtained of the 5 self-weaned patients. The prospectively weaned recipients (93 livers; 2 liver/kidney) had undergone transplantation under immunosuppression based on azathioprine (AZA, through 1979), cyclosporine (CsA, 1980-1989), or tacrolimus (TAC, 1989-1994). In patients on CsA or TAC based cocktails, the adjunct drugs were weaned first in the early part of the trial. Since 1994, the T cell-directed drugs were weaned first. Three of the 5 original self-weaned recipients remain well after drug-free intervals of 14 to 17 years. A fourth patient died in a vehicular accident after 11 years off immunosuppression, and the fifth patient underwent retransplantation because of hepatitis C infection after 9 drug-free years; their allografts had no histopathologic evidence of rejection. Eighteen (19%) of the 95 patients in the prospective series have been drug free for from 10 months to 4.8 years. In the total group, 18 (19%) have had biopsy proved acute rejection; 7 (7%) had a presumed acute rejection without biopsy; 37 (39%) are still weaning; and 12 (13%, all well) were withdrawn from the protocol at reduced immunosuppression because of noncompliance (n=8), recurrent PBC (n=2), pregnancy (n=1), and renal failure necessitating kidney transplantation (n=1). No patients were formally diagnosed with chronic rejection, but 3 (3%) were placed back on preexisting immunosuppression or switched from cyclosporine (CsA) to tacrolimus (TAC) because of histopathologic evidence of duct injury. Two patients with normal liver function died during the trial, both from complications of prior chronic immunosuppression. No grafts suffered permanent functional impairment and only one patient developed temporary jaundice. Long surviving liver transplant recipients are systematically overimmunosuppressed. Consequently, drug weaning, whether incomplete or complete, is an important management strategy providing it is done slowly under careful physician surveillance. Complete weaning from CsA-based regimens has been difficult. Disease recurrence during drug withdrawal was documented in 2 of 13 patients with PBC and could be a risk with other autoimmune disorders.

Assessing risk in liver transplantation. Special reference to the significance of a positive cytotoxic crossmatch

OBJECTIVE

The authors determined the impact of a positive cytotoxic crossmatch on the outcome of liver transplantation.

SUMMARY BACKGROUND DATA

Liver allografts rarely undergo hyperacute rejection, but transplants performed across a positive cytotoxic crossmatch tend to follow a different clinical course, with higher intraoperative blood use, postoperative graft dysfunction, and, in some cases, graft loss. How this affects overall graft survival has not been determined.

METHODS

The authors provide a retrospective analysis of 1520 liver transplants performed between November 1989 and December 1993, with a minimum follow-up of 1 year. All cases had a cytotoxic crossmatch using serum pretreated with dithiothreitol.

RESULTS

There were 1390 negative crossmatch and 130 positive crossmatch cases. There was no difference in overall graft survival, although early survival rates were lower in the positive crossmatch group, with the maximum difference at 6 months: 0.76 (95% confidence interval, 0.74-0.78) for a negative crossmatch versus 0.68 (95% confidence interval, 0.61-0.77) for a positive crossmatch. These differences become negligible by the 2-year mark. Using stepwise logistic regression, the authors identified seven variables independently associated with outcome: 1) donor age, 2) donor gender, 3) prior liver transplant, 4) medical urgency status, 5) ischemia time, 6) indication for transplantation, and 7) primary immunosuppressant.

CONCLUSIONS

The cytotoxic crossmatch is not statistically associated with overall graft survival after liver transplantation. However, early failure rates are higher in the positive crossmatch cases, a difference that disappears by the second year.

Computer-guided randomized concentration-controlled trials of tacrolimus in autoimmunity: multiple sclerosis and primary biliary cirrhosis

A randomized concentration-controlled clinical trial (RCCCT) is a trial design in which patients are randomized to predefined blood drug concentrations (low, medium, high). If the concentration ranges are sufficiently separated, this study design can reveal important blood concentration-response relations. Tacrolimus is a potent yet "infant" immunosuppressant for the treatment and prevention of graft rejection and has been shown to exhibit significant clinical activity in some immune-mediated disorders. A tacrolimus artificial intelligence modeling system (AIMS) was used to guide patient dosing to achieve target concentrations specified by the study protocols. In the Multiple Sclerosis study group, we were able to define a concentration range (0.3-0.7 ng/ml) that appeared to show efficacy and minimal tacrolimus toxicity. Patients randomized to the high zone (0.6-1.2 ng/ml) in the Primary Biliary Cirrhosis study group showed significant reduction (approximately 50%) in surrogate efficacy markers [aspartate aminotransferase (SGOT), alanine aminotransferase (SGPT)] compared with patients in the low zone (0.1-0.6 ng/ml). Therefore the RCCCT allowed the detection and delineation of clinically significant concentration-response relations in an ethical and efficient manner.

Role of pharmacokinetic-pharmacodynamic principles in rational and cost-effective drug development

An important goal of drug development is to define dose and concentration-response relationships for new drugs and biologics. Such critical information from controlled clinical trials can provide primary evidence of efficacy and safety and an informative database for devising dosing instructions for clinical use. This article describes applications of pharmacologic principles [pharmacokinetic-pharmacodynamic (PK-PD)] and modeling methods for drugs in which the evaluation process is guided by and/or identifies significant PK and/or PD variability in drug response. In the case of the recently registered immunosuppressive agent, tacrolimus, preclinical PK-PD in model systems can be used to rationally design safe and effective immunomodulatory dosing regimens for phase 1 clinical studies. Furthermore, a study design based on concentration control guided by a novel artificial intelligence modeling system (AIMS) can be efficiently applied to conduct randomized clinical trials in auto-immunity and to implement cost-effective therapeutic drug monitoring of tacrolimus and cyclosporine in clinical transplantation. In the case of a cardioselective beta-adrenergic blocking agent, betaxolol, marketed for essential hypertension, population PD modeling can be shown to be a more efficient method for estimating dose response compared with standard statistical tests. Using a sigmoid Emax PD model, only a fraction (40 of 300) of the randomized patients was needed to demonstrate dose response. Therefore, two methods, i.e., PD modeling of dose response and AIMS-guided dosing, can achieve significant cost benefits for drug developers, patient care, and the health care system.

Hepatic Retransplantation--an analysis of risk factors associated with outcome

Hepatic retransplantation is controversial because the results are inferior to primary transplants and organs are so scarce. To determine the factors that are associated with poor outcome within the first year following retransplantation, we performed a multivariate analysis, using stepwise logistic regression, of 418 hepatic retransplantations performed at a single institution from November 1987 to December 1993. The minimum follow-up was 1 year. Seven variables were found to be independently associated with subsequent graft failure (defined as either patient death or retransplantation): donor age (odds ratio 2.2 for each 10-year increase over age 45, 95% CI 1.3 to 3.7), female donor sex (odds ratio 1.7, 95% CI 1.05 to 2.7), recipient age (odds ratio 1.6 for each 10-year increase over age 45,95% CI 1.2 to 2.8), need for preoperative mechanical ventilation (odds ratio 1.8, 95% CI 1.1 to 2.9), pretransplant serum creatinine (odds ratio 1.24 for each increase of 1 mg/dl, 95% CI 1.1 to 1.4), pretransplant total serum bilirubin (odds ratio 1.4 for each 10-mg/dl increase over 15 mg/dl, 95% CI 1.1 to 1.8), and the primary immunosuppressant, using tacrolimus as the reference category (odds ratio for cyclosporine-based immunosuppression 3.9, 95% CI 2.3 to 6.8). Although not part of the logistic regression model, the timing of retransplantation was also found to be important, with the overall probability of failure increasing from 0.58 on day 0 to a peak of 0.8 on day 38 and decreasing slowly after that. The implications of these results regarding the appropriateness of retransplantation are discussed.

Performance of multi-layer feedforward neural networks to predict liver transplantation outcome

A novel multisolutional clustering and quantization (MCQ) algorithm has been developed that provides a flexible way to preprocess data. It was tested whether it would impact the neural network's performance favorably and whether the employment of the proposed algorithm would enable neural networks to handle missing data. This was assessed by comparing the performance of neural networks using a well-documented data set to predict outcome following liver transplantation. This new approach to data preprocessing leads to a statistically significant improvement in network performance when compared to simple linear scaling. The obtained results also showed that coding missing data as zeroes in combination with the MCQ algorithm, leads to a significant improvement in neural network performance on a data set containing missing values in 59.4% of cases when compared to replacement of missing values with either series means or medians.

Effect of donor age and sex on the outcome of liver transplantation

We correlated donor and recipient factors with graft outcome in 436 adult patients who underwent 462 liver transplants. Donor variables analyzed were age, gender, ABO blood group, cause of death, length of stay in the intensive care unit, use of pressors or pitressin, need for cardiopulmonary resuscitation, terminal serum transaminases, and ischemia time. Recipient variables analyzed were age, gender, primary diagnosis, history of previous liver transplant, ABO blood group, cytotoxic antibody crossmatch, United Network for Organ Sharing (UNOS) status, and waiting time (except for the cross-match results, they were all known at the time of the operation). The endpoint of the analysis was graft failure, defined as patient death or retransplantation. Using multivariate analysis, graft failure was significantly associated with donor age, donor gender, previous liver transplantation, and UNOS 4 status of the recipient. The effect of donor age became evident only when they were older than 45 years. Livers from female donors yielded significantly poorer results, with 2-year graft survival of female to male 55% (95% CI, 45% to 67%); female to female, 64% (95% CI, 54% to 77%); male to male, 72% (95% CI, 66% to 78%); and male to female, 78% (95% CI, 70% to 88%). The only donors identified as questionable for liver procurement were old (> or = 60 years) women in whom the adverse age and gender factors were at least additive. However, rather than discard even these livers, in the face of an organ shortage crisis, their individualized use is suggested with case reporting in a special category.

Effect of donor age and sex on the outcome of liver transplantation

We correlated donor and recipient factors with graft outcome in 436 adult patients who underwent 462 liver transplants. Donor variables analyzed were age, gender, ABO blood group, cause of death, length of stay in the intensive care unit, use of pressors or pitressin, need for cardiopulmonary resuscitation, terminal serum transaminases, and ischemia time. Recipient variables analyzed were age, gender, primary diagnosis, history of previous liver transplant, ABO blood group, cytotoxic antibody crossmatch, United Network for Organ Sharing (UNOS) status, and waiting time (except for the cross-match results, they were all known at the time of the operation). The endpoint of the analysis was graft failure, defined as patient death or retransplantation. Using multivariate analysis, graft failure was significantly associated with donor age, donor gender, previous liver transplantation, and UNOS 4 status of the recipient. The effect of donor age became evident only when they were older than 45 years. Livers from female donors yielded significantly poorer results, with 2-year graft survival of female to male 55% (95% CI, 45% to 67%); female to female, 64% (95% CI, 54% to 77%); male to male, 72% (95% CI, 66% to 78%); and male to female, 78% (95% CI, 70% to 88%). The only donors identified as questionable for liver procurement were old (> or = 60 years) women in whom the adverse age and gender factors were at least additive. However, rather than discard even these livers, in the face of an organ shortage crisis, their individualized use is suggested with case reporting in a special category.

Clinical pharmacokinetics of tacrolimus

Tacrolimus, a novel macrocyclic lactone with potent immunosuppressive properties, is currently available as an intravenous formulation and as a capsule for oral use, although other formulations are under investigation. Tacrolimus concentrations in biological fluids have been measured using a number of methods, which are reviewed and compared in the present article. The development of a simple, specific and sensitive assay method for measuring concentrations of tacrolimus is limited by the low absorptivity of the drug, low plasma and blood concentrations, and the presence of metabolites and other drugs which may interfere with the determination of tacrolimus concentrations. Currently, most of the pharmacokinetic data available for tacrolimus are based on an enzyme-linked immunosorbent assay method, which does not distinguish tacrolimus from its metabolites. The rate of absorption of tacrolimus is variable with peak blood or plasma concentrations being reached in 0.5 to 6 hours; approximately 25% of the oral dose is bioavailable. Tacrolimus is extensively bound to red blood cells, with a mean blood to plasma ratio of about 15; albumin and alpha 1-acid glycoprotein appear to primarily bind tacrolimus in plasma. Tacrolimus is completely metabolised prior to elimination. The mean disposition half-life is 12 hours and the total body clearance based on blood concentration is approximately 0.06 L/h/kg. The elimination of tacrolimus is decreased in the presence of liver impairment and in the presence of several drugs. Various factors that contribute to the large inter- and interindividual variability in the pharmacokinetics of tacrolimus are reviewed here. Because of this variability, the narrow therapeutic index of tacrolimus, and the potential for several drug interactions, monitoring of tacrolimus blood concentrations is useful for optimisation of therapy and dosage regimen design.

Building clinical classifiers using incomplete observations--a neural network ensemble for hepatoma detection in patients with cirrhosis

One objective of liver transplant evaluation is to identify patients that harbor a hepatoma, but standard screening techniques are not sensitive enough. We trained neural network ensembles to predict the presence of hepatoma in patients with cirrhosis, based on information collected at the time of transplant evaluation. Network architecture and training were modified to handle missing observations. Three ensembles were trained: ensemble A using the subset with no missing observations (528 patients); ensemble B using the complete set, which included missing observations (853 patients); and ensemble C using the smaller subset, originally with complete data, but after a fixed number of observations were deleted (i.e., made "missing"). Ensemble performance on testing sets was very good. The areas under the ROC curves were 0.91, 0.89, and 0.90, for ensembles A, B, and C, respectively. Neural networks can successfully perform this classification task, and strategies can be developed that allow use of incomplete observations.

Tacrolimus: a potential new treatment for autoimmune chronic active hepatitis: results of an open-label preliminary trial

Autoimmune chronic active hepatitis (CAH-A) is a chronic liver disease of unknown etiology that is believed to have an autoimmune pathogenesis. The disease is slowly progressive until hepatic failure and portal hypertension develop and either death or liver transplantation occur. Currently, the only widely recognized therapy is the administration of glucocorticoids, which have both anti-inflammatory and immunosuppressive actions. Many patients cannot tolerate such therapy because of the psychiatric, osteoporotic, and weight-enhancing actions of steroids. Tacrolimus (FK 506) is a new macrolide antibiotic that has an immunosuppressive activity that is estimated to be 10-200 times greater than that of cyclosporine. Because of its greater immunosuppressive activity, we have used it in the treatment of 21 patients with autoimmune chronic active hepatitis. Before each subject was treated, a liver biopsy and a panel of hematological, serological, and biochemical parameters were assessed. The Tacrolimus was administered orally at 12-h intervals, and the dose was controlled by monitoring plasma FK trough levels. After 3 months of therapy at an oral dose of 3 mg twice a day, having achieved a median blood level of 0.5 ng/ml, the serum ALT level was reduced by 80%, and the AST level was reduced by 70%. Modest change in the white blood cell count and platelet count were noted. The median BUN level increased from a level of 12 to 18 mg/dl, and the serum creatinine increased from 0.9 to 1.3 mg/dl. These preliminary data demonstrate that: 1) Tacrolimus can be used to successfully treat CAH-A; 2) the response of CAH-A to Tacrolimus treatment is rapid and sustained; and 3) a minor increase in the serum BUN and creatinine levels occurs as a consequence of Tacrolimus treatment. It is anticipated that with continued treatment for periods of 1-2 yr, the natural history of CAH-A will be changed such that hepatic failure and the requirement for liver transplantation may be averted.

Tacrolimus (FK 506), a treatment for primary sclerosing cholangitis: results of an open-label preliminary trial

Primary sclerosing cholangitis (PSC) is a chronic inflammatory disease of the liver that is characterized by progressive cholestasis and the development of secondary biliary cirrhosis. There is no widely recognized therapy for this disease, although anti-inflammatory agents (steroids), immunosuppressive agents (methotrexate), anti-fibrotics (colchicine), and choleretic agents (ursodeoxycholic acid) have been used in various small series. In the present study, Tacrolimus (FK 506), a new and powerful immunosuppressive macrolide antibiotic, has been used to treat 10 patients with PSC. Each subject had a liver biopsy, ERCP with visualization of the intra- and extrahepatic biliary tree, and a panel of hematological, serological, and biochemical laboratory tests before the initiation of the FK 506 therapy. The FK 506 was administered orally at 12-h intervals and was monitored by serial plasma FK 506 trough levels. After 360 days of treatment, the median serum bilirubin level was reduced by 75%, and the serum alkaline phosphatase was reduced by 70%. Moreover, the serum ALT and AST levels were reduced by 80 and 86%, respectively. No change in the serum level of BUN and creatinine levels occurred as a consequence of the FK 506 treatment. These data demonstrate that: 1) FK 506 can be used to treat PSC; 2) the response to FK 506 by patients with PSC is rapid; and, 3) no adverse effect on the serum BUN and creatinine levels was observed. It is anticipated that FK 506 will become an important agent for the treatment of patients with PSC because of its powerful immunosuppressive activity.

Weaning of immunosuppression in long-term liver transplant recipients

Seventy-two long-surviving liver transplant recipients were evaluated prospectively, including a baseline allograft biopsy for weaning off of immunosuppression. Thirteen were removed from candidacy because of chronic rejection (n = 4), hepatitis (n = 2), patient anxiety (n = 5), or lack of cooperation by the local physician (n = 2). The other 59, aged 12-68 years, had stepwise drug weaning with weekly or biweekly monitoring of liver function tests. Their original diagnoses were PBC (n = 9), HCC (n = 1), Wilson's disease (n = 4), hepatitides (n = 15), Laennec's cirrhosis (n = 1), biliary atresia (n = 16), cystic fibrosis (n = 1), hemochromatosis (n = 1), hepatic trauma (n = 1), alpha-1-antitrypsin deficiency (n = 9), and secondary biliary cirrhosis (n = 1). Most of the patients had complications of long-term immunosuppression, of which the most significant were renal dysfunction (n = 8), squamous cell carcinoma (n = 2) or verruca vulgaris of skin (n = 9), osteoporosis and/or arthritis (n = 12), obesity (n = 3), hypertension (n = 11), and opportunistic infections (n = 2). When azathioprine was a third drug, it was stopped first. Otherwise, weaning began with prednisone, using the results of corticotropin stimulation testing as a guide. If adrenal insufficiency was diagnosed, patients reduced to < 5 mg/day prednisone were considered off of steroids. The baseline agents (azathioprine, cyclosporine, or FK506) were then gradually reduced in monthly decrements. Complete weaning was accomplished in 16 patients (27.1%) with 3-19 months drug-free follow-up, is progressing in 28 (47.4%), and failed in 15 (25.4%) without graft losses or demonstrable loss of graft function from the rejections. This and our previous experience with self-weaned and other patients off of immunosuppression indicate that a significant percentage of appropriately selected long-surviving liver recipients can unknowingly achieve drug-free graft acceptance. Such attempts should not be contemplated until 5-10 years posttransplantation and then only with careful case selection, close monitoring, and prompt reinstitution of immunosuppression when necessary.

The current status of hepatic transplantation at the University of Pittsburgh

Tacrolimus is a more potent and satisfactory immunosuppressant than CyA for combination therapy with prednisone. In randomized trials comparing the 2 drugs, the ability of tacrolimus to rescue intractably rejecting grafts on the competing CyA arm allowed equalization of patient and graft survival on both arms when the intent-to-treat analytic methodology was applied. The ability of tacrolimus to systematically rescue the treatment failures of CyA suggested, as a matter of common sense, that it is the preferred baseline drug for hepatic transplantation. This conclusion was supported by analysis of secondary end points, including the ability to prevent rejection. Hepatic-intestinal, multivisceral and isolated intestinal transplantation became feasible on a practical basis only after the advent of tacrolimus. Nevertheless, better management strategies must be devised before intestinal transplantation, alone or with other abdominal viscera, will meet its potential. One such strategy is based on the discovery of the presence of previously unsuspected, low-level donor leukocyte chimerism in long-surviving allograft recipients. We believe that this chimerism is the essential explanation for the feasibility of organ transplantation and a link to the acquired neonatal tolerance demonstrated by Billingham, Brent and Medawar (32). The hematolymphopoietic chimerism in organ recipients explains why weaning to a drug-free state in selected long-term survivors is frequently feasible and particularly if the allograft is a liver. Weaning should never be attempted without a stepwise protocol and careful monitoring of graft function. Recognition of the natural chimerism that develops after whole organ transplantation has led to efforts to augment it with perioperative donor BM infusion. This procedure has been shown to be free of significant complications (including GVHD) in all kinds of whole organ recipients, including those given intestine. The prospects of clinical xenotransplantation must be evaluated in the same context of chimerism as that delineated for allotransplantation with the discovery of spontaneous chimerism. Before addressing chimerism-related questions in xenotransplantation, the additional barrier of the complement activation syndromes that cause hyperacute rejection will have to be surmounted. Although measures to effectively transplant xenografts have so far eluded us, the availability of the more potent drug, tacrolimus, and recognition of the seminal basis of allograft (or xenograft) acceptance via chimerism has inserted an element of reality into the largely wishful thinking that has been evident in discussions about the future of xenotransplantation.

Early death or retransplantation in adults after orthotopic liver transplantation. Can outcome be predicted?

Early, reliable outcome prediction after a liver transplant would help improve organ use by minimizing unnecessary retransplantations. At the same time, early intervention in those cases destined to fail may ameliorate the high morbidity and mortality associated with retransplantation. The purpose of this study was to analyze several parameters that have been identified in the past as being associated with patient and graft outcome, and to try to develop a model that would allow us to make predictions based on data available in the early postoperative period. A total of 148 patients were followed in a prospective, observational study. Graft failure was defined as patient death or retransplantation within 3 months of surgery. Preoperative variables studied included patient demographics, need for life support, presence of ascites, serum bilirubin, serum albumin, prothrombin time, serum creatinine, and the results of the cytotoxic crossmatch. During the first 5 postoperative days, standard measurements included serum transaminases, serum bilirubin, ketone body ratio, prothrombin time, factor V, and serum lactate. Oxygen consumption was measured shortly after surgery, once the patients had rewarmed to 36 degrees C. There were 131 successful transplants (88.5%) and 17 failures (11.5%). Most of the variables studied were found to be associated with outcome (by univariate analysis) at different points in the early postoperative period. However, receiver operating characteristic curve analysis showed that the predictive ability of even the best parameter was not adequate to make decisions on individual patients. Multivariate analysis, using stepwise logistic regression, yielded a model with an overall accuracy of 92.7%. Again, receiver operating characteristic curve analysis suggested that this model did not achieve the discriminating power needed for routine clinical use. We are still not able to accurately predict outcome in the early posttransplant period. We must be very careful when evaluating parameters, or scoring systems, that are said to accomplish this. It is especially important in this era of cost containment, with its renewed pressures to guide therapy based on our perceived understanding of a patient's future clinical course.

Predicting outcomes after liver transplantation. A connectionist approach

OBJECTIVE

The authors sought to train an artificial neural network to predict early outcomes after orthotopic liver transplantation.

SUMMARY BACKGROUND DATA

Reliable prediction of outcomes early after liver transplantation would help improve organ use and could have an impact on patient survival, but remains an elusive goal. Traditional multivariate models have failed to attain the sensitivity and specificity required for practical clinical use. Alternate approaches that can help us model clinical phenomena must be explored. One such approach is the use of artificial neural networks, or connectionist models. These are computation systems that process information in parallel, using large numbers of simple units, and excel in tasks involving pattern recognition. They are capable of adaptive learning and self-organization, and exhibit a high degree of fault tolerance.

METHODS

Ten feed-forward, back-propagation neural networks were trained to predict graft outcomes, using data from 155 adult liver transplants. The data included information that was available by the second postoperative day. Ten separate training and testing data subsets were prepared, using random sampling, and the ability of the different networks to predict outcomes successfully was evaluated using receiver operating characteristic (ROC) curve analysis.

RESULTS

Four of the networks showed perfect discrimination, with an area under the ROC curve (Az) of 1.0. Two other networks also had excellent performance, with an Az of 0.95. The sensitivity and specificity of the combined networks was 60% and 100%, respectively, when using an output neuron activation of 0.6 as the cutoff point to decide class membership. Lowering the cutoff point to 0.14 increased the sensitivity to 77%, and lowered the specificity to 96%.

CONCLUSIONS

These results are encouraging, especially when compared to the performance of more traditional multivariate models on the same data set. The robustness of neural networks, when confronted with noisy data generated by nonlinear processes, and their freedom from a priori assumptions regarding the data, make them promising tools with which to develop predictive clinical models.

Computer-guided concentration-controlled trials in autoimmune disorders

A randomized concentration-controlled clinical trial (RCCCT) is an alternate experimental design to the standard dose-controlled study. In a RCCCT, patients are randomly assigned to predefined plasma or blood drug concentration ranges (low, medium, and high). With the caveat that concentration ranges are sufficiently separated, this design should enhance the ability to discover important concentration response relationships. FK-506, a potent and promising immunosuppressive agent for prevention and treatment of graft rejection, has shown significant clinical activity in some immune-mediated disorders. To implement the RCCCT design, a novel FK-506 intelligent dosing system (IDS) was used to guide all doses to prospectively achieve the target concentration range specific in the study protocol. Patients enrolled in these trials suffered from a variety of autoimmune disorders, including multiple sclerosis, primary biliary cirrhosis, psoriasis, autoimmune chronic active hepatitis, and nephrotic syndrome. We observed excellent predictive performance of the IDS for all patients. The accuracy (mean prediction error) of the IDS was -0.022 ng/ml and the precision (standard deviation of the prediction error) was 0.119 ng/ml. Thus, the IDS is both accurate and reproducible for autoimmune patients. We conclude that the RCCCT design, guided by an accurate and precise IDS, is an informative and cost-effective approach for evaluation of efficacy and safety of effective but highly toxic agents.