Transplant plenary I


1

MELD Inflation: The Current Hepatocellular Carcinoma Exception Policy is Primarily Responsible for Steadily Increasing MELD Scores at the Time of Liver Transplant in All Regions of the U. S

Patrick G. Northup, Nicolas M. Intagliata, Neeral L. Shah, Curtis K. Argo

Gastroenterology and Hepatology, University of Virginia, Charlottesville, VA

With large increases in hepatocellular carcinoma (HCC) diagnoses in recent years and advances in HCC locoregional therapies, more HCC patients are awaiting liver transplantation than ever before. It was the aim of this study to determine if HCC exceptions are related to the increase in MELD scores on the U. S. liver transplant waiting list. Methods: Data from all adult, non-status one, initial transplant candidates who were listed for liver transplantation between January 2005 and December 2012 were analyzed. Because of significant changes in HCC exceptions from 2002-2004, these years were excluded from the analysis. Yearly trends in waiting list characteristics and transplantation rates were analyzed for statistical association with MELD exceptions, especially HCC exceptions, using multivariate generalized linear modeling. Regional variations in these associations were also analyzed. Results: During the study period there was a significant difference between the mean initial laboratory MELD score for candidates with any MELD exception (11.5) and no exception (18.3), p < 0.001. This gap widened at the time of waiting list removal (exception 12.6 vs. non-exception 21.9, p < 0.0001). Because of the extra exception points given to HCC candidates every three months, the mean score used for allocation significantly favored the HCC exception candidate (24.3 vs. 21.6, P < 0.0001). Candidates with HCC exceptions fared much better on the waiting list due to the lower lab-based MELD score compared to those without exceptions in mean days waiting (HCC 257 vs. non-HCC 420, P < 0.0001), transplantation rates (HCC 65.4% vs. non-HCC 44.2%, p < 0.0001)and waiting list death rates (HCC 13.5% vs. non-HCC 22.3%). The number of HCC-exceptions granted was directly and independently associated with the average MELD score of all candidates at the time of removal (p < 0.0001). Although the absolute number of exceptions differed, this trend was near universal in all UNOS regions over the same time period. Conclusions: The rates of HCC MELD exceptions and MELD scores at waiting list removal have increased steadily over the past 8 years. The MELD upgrades afforded to HCC exception candidates drive the increase in average MELD scores for all candidates in all UNOS regions. Alteration of the MELD upgrade amount or frequency for HCC patients should be considered to optimize equity on the transplant waiting list for all candidates.

Disclosures:

Patrick G. Northup - Grant/Research Support: Hemosonics, Bristol Meyer Squibb Neeral L. Shah - Grant/Research Support: Hemosonics

Curtis K. Argo - Consulting: Wellstat Diagnostics; Independent Contractor: Genentech/Roche

The following people have nothing to disclose: Nicolas M. Intagliata

2

Wait List Time Predicts Survival after Liver Transplantation for Hepatocellular Carcinoma: A Cohort Study in the UNOS Registry

Barry Schlansky1, Yiyi Chen2, Donald Austin2, Willscott E. Naugler1

1Gastroenterology & Hepatology; Oregon Health & Science University, Portland, OR; 2Public Health & Preventive Medicine, Oregon Health & Science University, Portland, OR

Purpose. Recipients of liver transplantation (LT) for hepatocellular carcinoma (HCC) harbor an 8-20% risk of HCC recurrence. Single-center studies suggest that an observation period after liver-directed HCC therapy may allow selection of candidates with reduced risk of post-LT HCC recurrence. We sought to determine whether wait list time after MELD prioritization for HCC predicts post-LT survival. Methods. In the UNOS registry, we selected 3 groups registered on the LT wait list from March 2005-March 2008: (1)patients receiving MELD prioritization for HCC, HCC+MP; (2) patients with HCC who did not receive MELD prioritization, HCC-MP; and (3) patients without HCC, NHCC. The primary exposure was time from initial MELD prioritization until LT, using the MELD status at LT. Recipients of LT were followed until death or censoring through October 2012.We used multiple Cox regression to evaluate the association of MELD status with post-LT survival, and we performed an intention-to-treat (ITT) analysis from wait list registration using timedependent models to estimate the implications of the association on wait list removal and LT benefit. Results. The median calculated MELD was 12 and MELD status 22 in the HCC+MP group, compared to 23 and 24 in the other 2 groups. One, 3 and 5-year post-LT survival were highest for the NHCC group (91%, 84%, 77%), intermediate for the HCC+MP group (92%, 81%, 73%), and lowest for the HCC-MP group (89%, 77%, 70%) (log-rank p < 0.01). Increasing MELD status at LT was independently associated with longer survival in the HCC+MP group (HR 0.84; 95% CI: 0.73-0.98); in contrast, increasing MELD status at LT was associated with shorter survival in the HCC-MP (HR 1.17, 95% CI: 1.02-1.33) and NHCC (HR: 1.20; 95% CI: 1.15-1.25) groups. In the ITT cohort, 75% of the HCC+MP group received LT and 23% experienced wait list removal or death, compared to 40% and 24% of the HCCMP group and 48% and 29% of the NHCC group. LT conferred a 67% mortality reduction in the HCC-MP group and a 61% reduction in the other 2 groups. Conclusion. Increasing time on the LT wait list after development of HCC predicts improved post-LT survival, likely due to the selection of candidates with favorable tumor biology and reduced HCC recurrence risk. In contrast, post-LT survival is associated with the severity of liver dysfunction rather than wait list time in candidates without MELD prioritization, regardless of HCC status. Since candidates with HCC have increased access to LT and lower post-LT survival, delaying LT for HCC candidates with preserved liver function may optimize ITT survival and equitable organ allocation across the overall pool of LT candidates.

Disclosures:

The following people have nothing to disclose: Barry Schlansky, Yiyi Chen, Donald Austin, Willscott E. Naugler

3

Twice-Daily Telaprevir in Combination with Peginterferon Alfa-2a/Ribavirin in Genotype 1 HCV Liver Transplant Recipients: Interim Week 16 Safety and Efficacy Results of the Prospective, Multicenter REFRESH Study

Kimberly Ann Brown1, Robert J. Fontana2, Mark W. Russo3, Josh Levitsky4, Eric M. Yoshida5, Hugo E. Vargas6, Mohammad Bsharat7, Raymond A. Ruoin7, Robert S. Brown8

1Henry Ford Hospital, Detroit, MI; 2University of Michigan, Ann Arbor, MI; 3Carolinas Medical Center, Charlotte, NC; 4Northwestern University, Chicago, IL; 5University of British Columbia, Chicago, IL; 6Mayo Clinc, Scottsdale, AZ; 7Vertex Pharmaceuticals Incorporated, Cambridge, MA; 8Columbia University, New York, NY

REFRESH, a 2-part, open-label, phase 2b study, is the first prospective, multicenter study assessing safety, efficacy, and pharmacokinetics of telaprevir (T) in combination with peginterferon alfa-2a (P) and ribavirin (R) in noncirrhotic liver transplant recipients with genotype 1 chronic hepatitis C virus (HCV) infection. We report interim wk 16 efficacy and safety data for 46 initial patients (pts). PR treatment-naϊve or experienced adults aged 18-65 yrs with genotype 1a or 1b chronic HCV and liver transplantation within 6 mos to 10 yrs of screening, on stable cyclosporine (CsA; n=7) or tacrolimus (TAC; n=39), were eligible for inclusion. Pts received T 1125 mg BID + P 180 μg/wk + R 600 mg/day (initial dose) for 12 wks, followed by 36 wks PR. CsA or TAC doses were held then reduced post-initial T ≈4-fold (min. dose, 25 mg) and ≈10-fold (min. dose, 0.5 mg), respectively. No PR lead-in was used. Enrolled pts (n=46) had mean (range) age of 57 yrs (43-65) and were predominantly male (83%), white (83%), genotype 1a (74%), and previously treated with PR (78%). Median baseline HCV RNA was 6.9 (4.1-7.6) log10IU/mL and fibrosis stages were F0/1 (16%), F2 (59%), and F3 (22%), with 4% missing. Key adverse events (AEs) through wk 16 were anemia (n=22, 8 mild, 11 moderate, 3 severe; 12 erythropoietin; 2 transfusions), rash (n=16, all mild), pruritus (n=10, 9 mild, 1 moderate), anorectal symptoms (n=19, 15 mild, 3 moderate, 1 severe), increased blood creatinine or renal failure (n=10, 5 mild, 5 moderate), and infections (n=10, 7 mild, 2 moderate, 1 serious). All study drugs were discontinued in 3 pts (n=1: herpes zoster; rhabdomyolysis; and muscle weakness, pruritus, edema, and elevated uric acid). T was discontinued in 1 pt with high TAC levels and increased creatinine wherein TAC was not held at T onset. No rejection, autoimmune hepatitis, or deaths reported. Available data for on-treatment HCV RNA levels are shown below. This interim analysis suggests that T/PR has an acceptable safety profile when given with CsA or TAC with manageable AEs in posttransplant recipients. Preliminary data show the majority of pts (53% and 60%) had undetectable HCV RNA at wks 4 and 12, respectively. No rejection, autoimmune hepatitis, or deaths were reported.

Variable, n (%)Wk 2(n=41)Wk 4 (n=43)Wk 12(n=30)
  1. *Evaluated using Roche Taqman HCV RNA assay v2.0 (LLOQ=25 IU/mL; HCV RNA values less than LLOQ were reported as HCV RNA not detected or less than 25 IU/mL HCV RNA detected)

HCV RNA not detected*4(9)23 (53)18(60)
<25 IU/mL, HCV RNA detected*15(37)9(21)3(10)
≥25 IU/mL*22 (54)11(26)3(10)
Discontinued006(20)

Disclosures:

Kimberly Ann Brown - Advisory Committees or Review Panels: CLDF, Merck, Salix, Gilead, Vertex, Novartis, Genentech, Gilead, Janssen, Novartis, Salix; Consulting: Blue Cross Transplant Centers, Salix; Grant/Research Support: CLDF, Gilead, Exalenz, CDC, BMS, Bayer-Onyx, Ikaria, Hyperion, Merck; Speaking and Teaching: Salix, Merck, Genentech, Gilead, CLDF, Vertex

Robert J. Fontana - Consulting: GlaxoSmithKline, tibotec; Grant/Research Support: Gilead, vertex, Ocera

Mark W. Russo - Grant/Research Support: Vertex, Merck; Speaking and Teaching: Vertex, Gilead, BMS

Eric M. Yoshida - Advisory Committees or Review Panels: Hoffman LaRoche, Gilead Sciences Inc, Vertex Inc; Grant/Research Support: Cangene Corporation, Hoffman LaRoche, Merck Inc/Schering Plough, Pfizer Inc, Norvartis Inc, Vertex Inc, Jannsen Inc, Gilead Sciences Inc, Boeringher Ingleheim Inc, Abbie (formerly Abbott Laboratories), Astellas; Speaking and Teaching: Gilead Sciences Inc, Cangene Corporation, Vertex Inc, Merck Inc

Hugo E. Vargas - Advisory Committees or Review Panels: Eisai; Grant/Research Support: Merck, Gilead, Idenix, Novartis, Vertex, Janssen, Bristol Myers, Ikaria, Abbott

Mohammad Bsharat - Employment: vertex Inc; Stock Shareholder: vertex Inc

Raymond A. Rubin - Employment: Vertex Pharmaceuticals

Robert S. Brown - Consulting: Salix, Janssen, Vertex; Grant/Research Support: Gilead, Merck, Vertex, AbbVie, Salix, Janssen, BI; Speaking and Teaching: Genentech, Gilead, Merck

The following people have nothing to disclose: Josh Levitsky

4

Molecular Integrated Approach for Early Biomarkers of HCC Development in Hepatitis C Infected Patients

Daniel Maluf1, Ricardo C. Gehrau1, Abdullah M. Al-Osaimi2, Curtis K. Argo2, Stephen H. Caldwell2, Valeria Mas1

1Surgery, UVA, Cahrlottesviile, VA; 2Medicine, UVA, Cahrlottesviile, VA

Background: Hepatocellular carcinoma (HCC) incidence increases in chronic hepatitis C (HCV) cirrhosis while its surveillance lack accuracy. Integrative study of epigenetics, gene expression, and proteomics in cirrhotic tissue (non-HCC) was performed to identify early mechanistics events associated with HCC development. Methods: The study included tissue biopsy samples from 76 livers including Normal (n = 20), HCV-cirrhosis without (woHCC =16) and with (wHCC = 20) HCC, and HCC (tumor tissue n = 20) together with 20 plasma samples (woHCC =10; wHCC =10). DNA and total RNA were isolated and used for Illumina GoldenGate Methylation BeadArray Cancer Panel I and gene expression microarray hybridization, respectively. Proteomic analyses were assayed by LC-MS/MS. Gene expression and methylation comparison analyses (wHCC vs. woHCC) were fit using two sample t-test. The p-values were adjusted for multiple comparisons by calculating the false discovery rate using the q-value method by controlling a FDR < 0.2.A p < 0.01 was considered significant. Spearman's rank correlation was used for methylation/gene expression comparison. Ontology and pathway analyses were performed using IPA tool. Results: Demographic and clinical parameters demonstrated no significant differences between study groups. Normal to HCV-cirrhosis plus HCC progression analysis identified 501 CpG sites with significant differential methylation (235 hypermethylated, 266 hypomethylated). Thirty hypermethylated CpG sites in promoter regions of tumor suppressor genes associated with HCC development. Top 10 hypermethylated genes included ESR1, MFAP4, and HOXB13 as most significant. ESR1 gene demonstrated top significant negative correlation between methylation percent and gene expression level (−0.50). DNA methylation and expression rates of several genes involved in HCC development (CDNK2A, FLT1, HGF, HPSE, MGMT, PDGFRB, and PYCARD) also significantly correlated with a negative trend indicating downregulation in HCV-cirrhosis with HCC. Pathways and expression trend analysis associated those genes with cancer and malignancies development, liver detoxification, and hepatic fibrosis/wound repair. Proteomic integration analysis revealed ESR1 as top signaling modulator of significant differentially expressed plasma proteins associated with HCC. Markers were validated using methyLight assay. Conclusions: Integration of epigenetic/transcriptomic/proteomic data provided affected molecular pathways critical for HCC development in HCV-cirrhotic tissue. Furthermore, different DNA methylation patterns appear to be potential biomarkers for HCC surveillance in high risk HCV cirrhotic patients.

Disclosures:

Abdullah M. Al-Osaimi - Advisory Committees or Review Panels: Vertex Pharmaceuticals, Merck; Grant/Research Support: Merck (previosuly Schering/Plough), GlaxoSmithKline, Gilead, Vertex Pharmaceuticals, Hoffman-LaRoche, Vital Therapies Inc, Hyperion Therapeutics, Inc, Bayer HealthCare Pharmaceuticals, Inc., Bristol-Myers Squibb Pharmaceutical Research Institute, HemoLife Medical, Inc., Ortho Biotech, Inc., Pharmasset, Octapharma, Novartis, Sanof Aventis, Bohrenger Inglheim; Speaking and Teaching: Vertex, Gilead

Curtis K. Argo - Consulting: Wellstat Diagnostics; Independent Contractor: Genentech/Roche

Stephen H. Caldwell - Advisory Committees or Review Panels: Vital Therapy; Consulting: Wellstat diagnostics; Grant/Research Support: Hemosonics, Gilead Sciences

The following people have nothing to disclose: Daniel Maluf, Ricardo C. Gehrau, Valeria Mas

5

Medication Tradeoffs, Non-Adherence, and Clinical Outcomes among Liver Transplant Recipients

Marina Serper1, 2, Kamila Przytula3, Rachel E. Patzer4, Daniela Ladner2, Michael S. Wolf2,3

1Division of Gastroenterology and Hepatology, Northwestern University, Chicago, IL; 2Northwestern University Transplant Outcomes Research Collaborative, Northwestern University, Chicago, IL; 3Health Literacy and Learning Program, Division of General Internal Medicine and Geriatrics, Northwestern University, Chicago, IL; 4Emory Transplant Center, Emory University, Atlanta, GA

Purpose: Patients with financial hardship may have to make tradeoffs between meeting basic needs (i. e. buying food) and buying medication. We aimed to examine the prevalence of these ‘medication tradeoffs' and their relationship to medication non-adherence and clinical outcomes among liver transplant (LT) recipients. Methods: We performed a cross-sectional study of 105 LT recipients at two transplant centers in Chicago, IL and Atlanta, GA between 2011 and 2012.In-person interviews were conducted to obtain baseline demographics, financial information, health literacy, medication tradeoffs (4 item questionnaire), and medication adherence (self-report). Clinical data were obtained from the medical record. Descriptive statistics, Chi-squared, t-tests, and various regression methods (linear, logistic, negative binomial) were used for analysis. Models were adjusted for demographics, insurance type, income, education, and health literacy. Results: Mean age of participants was 57±13, 59% were male and 81% non-Hispanic white. A total of 25% did not complete high school, 16% had income <$20K per year, and 12% were below the poverty level. The majority of participants (73%) were unemployed. A total of 73% had private insurance, 19% had Medicare or Medicaid, and 3% were self-pay. The median time since transplant was 20 months (IQR 9, 59) and average number of medications was 11 ± 4.A total of 18% of participants reported having medication tradeoffs; 11% reported an inability to buy a prescription due to cost and 11% reported having to choose between buying medicine and buying food. The prevalence of medication non-adherence by self-report was 23%. In bivariate analysis, race, insurance type, income, health literacy, and education were all significantly associated with medication tradeoffs (all p <. 05), while in multivariate models, lower income and limited literacy remained independently associated with medication tradeoffs (all p<. 05). Patients who had to make medication tradeoffs were less likely to report non-adherence (56% vs. 44%, p=. 05). In multivariate analysis, medication tradeoffs were independently associated with number of readmissions (IRR 2.0, 95%CI 1.1-3.7, p=. 02), while self-reported nonadherence was not. Conclusions: The prevalence of medication tradeoffs was significant and independently associated with post LT readmissions, while self-reported medication non-adherence was not. Despite the clinical significance, these patient choices may go undetected in routine practice. Assessing medication tradeoffs with the measure employed in this study could serve as simple clinical tool to identify patients at risk for skipping medications post LT.

Disclosures:

The following people have nothing to disclose: Marina Serper, Kamila Przytula, Rachel E. Patzer, Daniela Ladner, Michael S. Wolf

6

Magnitude of survival benefit after Liver Transplantation (LT) for metastatic neuroendocrine tumors (NET) of gastroentero-pancreatic (GEP) origin

Vincenzo Mazzaferro1, Carlo Sposito1, Rosalba Miceli4, Sherrie Bhoori1, Tiziana Camerini4, Jorgelina Coppa1, Enrico Regalia1, Carlo Battiston1, Marco A. Bongini1, Davide Citterio1, Carlo Spreafico3, Federica Brunero4, Filippo de Braud2, Luigi Mariani4

1Surgery & Hepatology, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, Milan, Italy; 2Medical Oncology, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, Milan, Italy; 3Radiology, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, Milan, Italy; 4Trial Office & Biomedical Statistics, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, Milan, Italy

LT for hepatic metastases from GEP-NET remains controversial due to lack of accepted selection criteria and competitive survival achieved with non-transplant options. Assessment of transplant-associated survival benefit (TSB) has been determined in two prospective cohorts of patients with similar burden of liver mets from GEP-NET allocated or not to LT. Material and methods During a 13-yr period 215 consecutive patients with liver mets from GEP-NET were referred for LT in a single Institution. Out of 110 pts considered eligible according to pre-determined criteria, 38 underwent LT after resection of the primary tumor (Group 1) while 72 with comparable tumor burden after removal of the primary tumor were treated with non-transplant options, due to age, compliance and co-morbidities (Group 2). Out of the remaining 115 patients, 55 underwent curative primary+liver resections (Group 3) while 50 received medical treatment only (Group 4). Overall survival (OS) was compared according to group allocation through multivariate Cox models adjusted for a propensity score (PS) that considered differences in covariates at baseline. The difference between Group 1 and Group 2 in mean survival time (DMST) at 5 and 10 yrs was considered the TSB. Pseudo-values method was applied, DMST was estimated by generalized estimating equation (GEE) models, with treatment allocation as the only independent variable (univariate model) or after PS adjustment (multivariate model). Results Median follow-up was 95 months. Baseline characteristics between Group 1 and 2 statistically differed for age but not for tumor burden, grading and stage. Cox models adjusted for PS demonstrated a significant survival advantage of Group 1 vs. Group 2, 3 and 4 (HR=5.54, 3.38, and 8.32 respectively; p=0.012). Five and 10-year OS in Group 1 was 96.8% and 92.2% vs. 59.3 and 37.7% in Group 2 (p<0.0001). After adjustment for PS, allocation to Group 1 was significantly associated with increased survival at multivariate analysis (HR=5.54, 95CI: 1.24-27.74, p=0.012). At univariate GEE analysis 5-yr and 10-yr TSB associated with allocation to transplant strategy was 11.6 and 42.4 months (p<0.001) while at multivariate analysis was 9.98 and 29.5 months respectively (p<0.001). Conclusions Liver transplantation for liver mets from GEP-NET may provide prolonged survival when restrictive criteria are applied. In patients with comparable tumor burden survival rates after LT significantly ameliorated with respect to that associated with surgical/locoregional/systemic treatments. Even after appropriate adjustments, a TSB approaching 30 months at 10 yrs may justify enlisting these patients as MELD exceptions.

Disclosures:

Vincenzo Mazzaferro - Advisory Committees or Review Panels: Bayer; Grant/Research Support: Nordion; Speaking and Teaching: Merck Serono S. p. A.

Sherrie Bhoori - Speaking and Teaching: Bayer, Nordion

The following people have nothing to disclose: Carlo Sposito, Rosalba Miceli, Tiziana Camerini, Jorgelina Coppa, Enrico Regalia, Carlo Battiston, Marco A. Bongini, Davide Citterio, Carlo Spreafico, Federica Brunero, Filippo de Braud, Luigi Mariani

Ancillary