Citations

Abstract Background The disease severity index (DSI) for inflammatory bowel disease (IBD) combines measures of disease phenotype, inflammatory activity, and patient-reported outcomes. We aimed to validate the DSI and assess its utility in predicting a complicated IBD course. Methods A multicenter cohort of adults with IBD was recruited. Intraclass correlation coefficients (ICCs) and weighted Kappa assessed inter-rater reliability. Cronbach’s alpha measured internal consistency of DSI items. Spearman’s rank correlations compared the DSI with endoscopic indices, symptom indices, quality of life, and disability. A subgroup was followed for 24 months to assess for a complicated IBD course. Area under the receiver operating characteristics curve (AUROC) and multivariable logistic regression assessed the utility of the DSI in predicting disease progression. Results Three hundred and sixty-nine participants were included (Crohn’s disease [CD], n = 230; female, n = 194; mean age, 46 years [SD, 15]; median disease duration, 11 years [interquartile range, 5-21]), of which 171 (CD, n = 99; ulcerative colitis [UC], n = 72) were followed prospectively. The DSI showed inter-rater reliability for CD (ICC 0.93, n = 65) and UC (ICC 0.97, n = 33). The DSI items demonstrated inter-rater agreement (Kappa > 0.4) and internal consistency (CD, α > 0.59; UC, α > 0.75). The DSI was significantly associated with endoscopic activity (CDn=141, r = 0.65, P < .001; UCn=105, r = 0.80, P < .001), symptoms (CDn=159, r = 0.69, P < .001; UCn=132, r = 0.58, P < .001), quality of life (CDn=198, r = −0.59, P < .001; UCn=128, r = −0.68, P < .001), and disability (CDn=83, r = −0.67, P < .001; UCn=52, r = −0.74, P < .001). A DSI of 23 best predicted a complicated IBD course (AUROC = 0.82, P < .001) and was associated with this end point on multivariable analyses (aOR, 9.20; 95% confidence interval, 3.32-25.49). Conclusions The DSI reliably encapsulates factors contributing to disease severity and accurately prognosticates the longitudinal IBD course.

Objective This pilot study seeks to identify serum immune signatures across clinical stages of patients with chronic pancreatitis (CP). Methods We performed a cross-sectional analysis of prospectively collected serum samples from the PROspective Evaluation of Chronic Pancreatitis for EpidEmiologic and Translation StuDies-study. CP subjects were categorised into three clinical stages based on the presence/absence of metabolic complications: (1) CP with no diabetes and exocrine pancreatic dysfunction (EPD), (2) CP with either diabetes or EPD, and (3) CP with diabetes and EPD. Blinded samples were analysed using an 80-plex Luminex assay of cytokines/chemokines/adhesion molecules. Group and pairwise comparisons were performed to characterise immune signatures across CP subgroups. Results A total of 135 CP subjects (evenly distributed between clinical stages) and 50 controls were studied. Interleukin-6 (IL-6), interleukin-8 (IL-8), and soluble intercellular adhesion molecule 1 (sICAM-1) were significantly elevated in CP subjects compared to controls. The levels of IL-6 and IL-8 increased with advancing disease stages, with the highest levels observed in CP with diabetes and EPD (clinical stage 3). Furthermore, hepatocyte growth factor and macrophage-derived chemokine were significantly increased in clinical stage 3 compared to controls. Conclusion Our study reveals a progressive elevation in pro-inflammatory cytokines and chemokines with advancing clinical stages of CP. These findings indicate potential targets for the development of disease-modifying interventions.

Scoring systems designed to improve the accuracy of prognostication in chronic lymphocytic leukemia (CLL) usually rely on discretized/dichotomic values of the clinical and biological variables. Here we analyzed the immunophenotypic and (immuno)genetic profiles of 2,243 CLL patients with Rai stage 0-I-II, by applying unsupervised machine learning methods, elaborating prognostic factors as continuous variables, to identify novel relationships and interactions likely missed in conventional hierarchical models. The study included an internal training cohort (n=863) and two external validation cohorts provided from Cardiff University/Brighton Medical School (n=455) and the Mayo Clinic (Rochester, MN; n=925). Primary endpoint was Time to First Treatment (TTFT). Laboratory-based markers evaluated in the training cohort were: surface antigen expression by flow cytometry (as % of positive cells) of CD20, FMC7, CD49d, CD49c, CD38, CD23, CD43, CD22, ZAP-70; chromosomal aberrations (% of nuclei with abnormal signal) del13q, tris12, del11q and del17p,; mutational status of TP53 by NGS as % VAF; IGHV mutational status as % mutation. 420/863 cases received treatment (median 46 months). Univariate Cox regression identified FISH del11q, del17p and tris12, TP53 and IGHV mutations, CD38 and CD49d expression, as features associated with TTFT (p-value <0.001). An unsupervised k-means algorithm partitioned cases into 6 clusters (C1-C6), as selected by the elbow method, and centroid analysis evaluated which feature most contributed to each cluster (Panel A): C1 (n=275) was heavily IGHV-mutated (mutation range 4.5-22.0%) with low representation of all other features; C2 (n=208) was mostly IGHV-unmutated/intermediate (up to 4.5%) in the absence of other features; C3 (n=169) were CD49d-expressing cases (>50% expression), with low representation of other features, except CD38; C4 (n=127) were cases with trisomy 12 (>28%), also expressing CD49d and CD38; C5 (n=48) were cases with highly clonal del11q (44-98%) and IGHV-unmutated, mutually exclusive with TP53 disruption; C6 (n=34) contained TP53-disrupted cases with high mutation burden (VAF 36-97%) and/or del17p (range 40-96%), with low importance of all the other features. Of note, alternative use of either TP53 mutation or del17p as defining features of cluster 6 did not result in changes of the classification accuracy. Clinically, clusters 3-4-5-6 showed median TTFT of 26, 22, 9, 5 months respectively whereas median TTFT was not reached for clusters 1-2. Hierarchical agglomerative clustering aggregated the 6 clusters in 3 major risk tiers: high (C5-6), intermediate (C2-3-4) and low (C1), with median TTFT of 7, 45 and not reached, respectively (Panel B). Multivariate Cox analysis (MVA) with Rai staging demonstrated that the 3-tier risk score contributed significantly and independently to TTFT estimate (p<0.0001; Harrel's c-index=0.71). To classify novel patients, we designed a hierarchical algorithm, based on each clusters' estimated risk and according to calculated cut-off, i.e. C6, del17p/ TP53>36%; C5, del11q>43%; C4, tris12>28%; C3, CD49d>50%; C2, IGHV≤4.5%; C1, the remaining patients with IGHV>4.5%. This approach was highly robust by an internal training-validation bootstrap (cluster classification accuracy = 0.96±0.02). External samples from two independent cohorts were classified into the six clusters using the cut-offs defined by the hierarchical classifier, and then aggregated into the 3-risk tiers. The classifier performed robustly in both cohorts, again adding independent prognostic information to Rai stage in MVA, with p<0.0001 and c-index of 0.76 for Cardiff/Brighton cohort and p<0.0001 and c-index of 0.74 for Mayo cohort. In conclusion, our machine-learning-driven, laboratory-based classification identifies clusters of patients who at higher risk of requiring early treatment, independent of clinical staging, and may help to better identify patients who may benefit from early treatment or more frequent disease surveillance. For some features, TP53 or IGHV mutational status, our unsupervised approach selected cut-offs not recapitulating the canonical ones, envisioning different biological activities in the TTFT setting. For example, TP53 mutations were clinically relevant only if present in the vast majority of the clone , suggesting the involvement of specific activities of the TP53 mutant in this context.

Background Diffuse large B cell lymphoma (DLBCL) is the most common aggressive lymphoid malignancy in adults. Though standard immunochemotherapy regimens can result in clinical remission and cure in a majority of patients, approximately 30% of patients are primary treatment resistant or eventually relapse (rrDLBCL), and their prognosis is very poor. Recent molecular classification by large scale genomic and transcriptomic profiling of newly diagnosed DLBCL tumors has resulted in a deeper understanding of disease drivers, however, the molecular heterogeneity of rrDLBCL is yet to be fully explored. Therefore, in this study, we used an unsupervised approach to better define transcriptional and genetic programs in rrDLBCL samples hypothesizing that molecular stratification within this high-risk group may lead to a better understanding of this disease, which remains an unmet need, and a more personalized therapeutic approach. Methods Data used in this study included RNA (n=143) and whole exome (n=126) sequencing data from available FFPE tumor samples at the time of a relapse (any line of treatment, r1-r10 relapse timepoints included in analysis, one per patient), consented to the Molecular Epidemiology Resource (n=61), banked in the Mayo Lymphoma Biobank (n=50), or consented to the CC-122-ST-001 clinical trial (n=32, NCT01421524). Unsupervised clustering was performed on protein coding gene expression values using the non-negative matrix factorization (NMF) approach. NMF was carried out by performing 200 runs on cluster sizes from 2 to 6. Cell of origin (COO) was determined using the method described by Reddy et al. The tumor microenvironment was analyzed using CIBERSORTx and Lymphoma Microenvironment Classification (LME). Genetic classification was done using LymphGen. Results In order to identify unique transcriptional programs in rrDLBCL, we implemented NMF and consensus clustering to define an ideal quantity of stable patient clusters. We identified 4 rrDLBCL patient clusters (rrC1-rrC4) as the most stable solution, based on consensus clustering selection criteria of iterative NMF runs. While relapse timepoint was not associated with individual clusters, we did observe significant differences in tumor characteristics between the clusters. COO classification found that rrC1 and rrC4 were enriched for GCB-DLBCL while rrC3 was enriched for ABC-DLBCL classification (P < .001). To explore the biologic programs in each cluster, we used pathFindR, which leverages protein-protein interaction networks to identify disease related pathways, rrC1 and rrC4 were defined by pathways involved in cell cycle and mismatch repair; oxidative phosphorylation in rrC2; and MAPK and NF-kB in rrC3. Next, LME classification was done to define patterns in the tumor microenvironment and we found a significant difference between groups (P < .001), with rrC2 and rrC4 tumors enriched for a depleted TME, while rrC1 and rrC3 had a more immune rich or inflammatory TME. This was supported by CIBERSORTx analysis identifying significant differences (P > .001) in abundance of CD8 T cells, memory CD4 T cells, and M1 macrophages between clusters, with rrC4 having the lowest abundance. We then explored genetic patterns in each cluster. Gene level mutation enrichment analysis showed BCL7A mutation exclusively in C4 (6/6, P = .03). The clusters were not associated with LymphGen classification. Lastly, our recently published high risk signature classification that is associated with early clinical failure and is characterized by metabolic dysregulation, a depleted TME, T53 alterations, and poor outcome, was significantly different between the groups (P = .03) with enrichment for high-risk cases in rrC4. In summary, we show for the first time that rrDLBCL patients can be classified into four gene expression clusters that are associated with distinct pathway, TME, and genetic programs. These clusters should now be tested to learn if they can help select patients for newer therapies for rrDLBCL such as CAR-T and bispecific antibodies.

Presence of deletion17p (del17p) in frontline (1L) and relapsed/refractory (R/R) CLL has been associated with shorter time to treatment initiation, early relapse after therapy and inferior overall survival (OS) (Mato et.al., Haemotologica 2022). However, there exists notable clinical heterogeneity in terms of outcome within this group (Tam et.al., Blood 2009, Stephens et.al; AJH 2015). Previous reports in 1L and R/R CLL have identified percentage of del17p nuclei by fluorescence in situ hybridization (FISH), having TP53 mutations on the second allele, copy number alterations and complex karyotype (CK) as additional risk factors within the del17p population (Yu et al., CCR 2017; Baliakas et al., Blood 2019). Here we profile a cohort of wait and watch (W&W, n=14), 1L treated (n=36) and R/R CLL (n=3) patients by whole genome sequencing (WGS) and RNA sequencing (RNAseq) to gain insights into genomic factors and gene expression pathways contributing to the high-risk nature of del17p across the spectrum of CLL patients. Using the Mayo Clinic CLL Database, we identified 53 patients with del17p (identified by FISH testing in routine clinical practice at diagnosis) who had stored peripheral blood mononuclear cells (PBMCs) available. RNA and DNA from matched peripheral blood (tumor) and DNA from CD19- (germline) samples at diagnosis (n=50) and relapse (n=3) was extracted using Qiagen RNeasy (RNA) and Purene (DNA) kits and subjected to RNAseq and WGS respectively. Additional FISH assays (del11q, trisomy 12, del13q) and CK by chromosomal banding analysis (CBA) were performed. Baseline clinical characteristics, time to first treatment, types of treatment, and survival were abstracted from the CLL Database. The Mayo Clinic IRB approved this study. Copy number analysis identified deletions in 3p, 4p, 8p, 9p, 9q and gains in 2p, 8q, 15q and 17q chromosomal lesions occurring in patients. We explored the association of clinical and genetic factors, and genomic instability signatures with survival to identify high-risk features in del17p patients. Both total number of deletions by WGS and number of chromosomal aberrations by CBA as a continuous variable were significantly associated with unfavorable OS in the entire cohort (n=53) in a univariate analysis (Hazard ratio, HR= 2.29, p= 0.00054 and HR= 2.33, p=0.0054 respectively). However, there was no significant association of number of gained genes by WGS or aneuploidy scores with survival (HR=0.74, p=0.37). In patients with CK data (n=41), deletion burden, but not gains, were highly correlated to chromosomal aberrations by CBA. Using cutpointr package for identifying deletion threshold corresponding to CK≥ 3, we defined patients with deletion burden ≥ 514 as enhanced-high risk (EHR; median OS = 4.83 years) compared to the group with high-risk (HR; median OS=9.94 years) (p=0.025). There was no significant difference in number of deleted genes between WW and treated patients. The EHR group is enriched in patients with high clonal cell fraction of del17p by FISH, unmutated IGHV and CK≥ 3. The EHR group remains significant in a multi-variable analysis with IGHV-mutation status, ZAP70, and age (HR: 5.16, 95% CI: 1.04 - 25.6, p=0.04). Mutational analysis identified 35 known CLL driver mutations with TP53 (86%), ATM (15%), NOTCH1(13%), MGA (13%), CHD2 and MED12(11%) as the top mutated genes. Total number of driver genes mutated was not significantly different between EHR and HR subgroups, but the EHR subset was significantly enriched with mutations in NOTCH1 (21%), and MGA (21%) as compared to no mutations in the HR group in these genes. Deleted genes (≥ 514) was prognostic in bi-allelic TP53 aberration patients as well (p=0.021) Differential expression gene-set enrichment analysis indicated a significant enrichment of DNA repair, cellular response to DNA damage and downregulation of inflammatory responses in the EHR group. CK is a known prognostic marker which includes both chromosomal losses and gains. In this small cohort of del17p we were able to show the prognostic value of deletions only, but not gains, as captured by WGS. Using genome wide sequencing, we identify increasing genomic deletions as a feature of enhanced-high risk del17p. While deletion burden cut-offs identified here are specific to our research method and require further validation in additional independent cohorts, EHR subgroup remained significant after adjusting for other known prognostic variables .

Introduction Atrial fibrillation/flutter (Afib) contributes independently to the hematopoietic cell transplant comorbidity index (HCTCI) and is a known complication of allogeneic hematopoietic transplantation (AlloHCT) which has been linked to inferior outcomes, particularly in older patients. An electrocardiogram (ECG)-based artificial intelligence (AI) algorithm to predict Afib and cardiac contractile dysfunction risk has been validated (Lancet 394:861, 2019). We hypothesized that assessment of pre-transplant Afib risk using the AI algorithm may predict future development of Afib and help identify patients at higher risk for this complication. Methods 191 consecutive MDS patients undergoing AlloHCT with curative intent using myeloablative (28.3%) or reduced intensity conditioning (62.3%) in the Mayo Clinic Comprehensive Cancer Center (Rochester, Jacksonville, Phoenix) between November 2015 and January 2022 were included in this retrospective analysis. Primary outcome was Afib within 100 days and 365 days of transplantation, and we also assessed the outcomes of Death and Relapse. Revised IPSS (IPSS-R) and HCTCI scores were included for all patients. Elevated Afib risk score using the AI algorithm was calculated from pre-transplant ECG, performed during vital organ testing at AlloHCT clearance. In addition to left ventricular (LV) ejection fraction (EF) we evaluated detailed echocardiographic (Echo) parameters including LV volume and mass indices, left atrial volume index (LAVI), maximum trans mitral flow velocities during LV filling (E velocity) and during the diastolic phase due to atrial contraction (A velocity), the velocity of the annulus away from the apex during the protodiastolic phase (e' velocity), E/A and E/e' ratios, the presence of valvular disease, and right ventricular systolic pressure (RVSP). We performed an analysis using a Cox proportional hazards regression models (adjusting for age and sex) to evaluate associations with Afib risk, Relapse, and Overall Survival. Results The median age was 64 years (18-75), 58.6% were male, 6.9% non-white race, and 3.7% Hispanic/Latino ethnicity. Most patients had an HCTCI score of 0-3 (76.3%). GVHD prophylaxis was predominantly with calcineurin inhibitor-based regimes plus/minus methotrexate or mycophenolate mofetil (83.8%). 5.2% of patients had a history of Afib before transplant. Non-relapse mortality was 7.9% at 100 days, and 18.3% at 365 days. N=25 patients (13.1%) developed Afib within 100 days, and n=28 (14.7%) within 1-year after AlloHCT. On multivariable analysis, Age [Hazard Ratio (HR) 2.83, 95% Confidence Interval (CI) 1.36-5.89, p=0.0055] and Echo parameters [including LAVI (HR 2.10, CI 1.23-3.58, p=0.0067), E velocity (HR 3.28, CI 1.29-8.80, p=0.013), E/A ratio (HR 3.74, CI 1.75-7.97, p=0.0006), E/e ' ratio (HR 1.81, CI 1.07-3.07, p=0.026) and RVSP (HR 1.68, CI 1.04-2.70, p=0.032)] were significantly associated with Afib risk; elevated AI risk score was significant in the univariate model (HR 2.74, CI 1.15-6.53, p=0.023) but was borderline in the multivariate model (HR 2.34, CI 0.98-5.60, p=0.056) ( Table 1). Based on previous observations we evaluated the association of AI and Echo parameters with Relapse and Death after AlloHCT ( Table 1). Age (HR 1.32, CI 1.01-1.71, p=0.041) and IPSS-R (HR 1.20, CI 1.08-1.33, p=0.0009) were significantly associated with Death after AlloHCT, while HCTCI (HR 1.10, CI 1.00-1.22, p=0.052) and higher LV EF (HR 0.63, CI 0.38-1.06, p=0.084) were borderline. For Relapse, only IPSS-R (HR 1.52, CI 1.31-1.75, p=1.22E-08) was significant, while increase Echo E velocity was of borderline significance (HR 0.45, CI 0.20-1.02, p=0.057). Conclusions Afib is a common post-AlloHCT complication and is more frequent with increasing Age. Echo parameters (specifically E/A ratio) predict significantly higher Afib risk, while the baseline ECG AI algorithm was of borderline significance. The selection of patients with lower HCTCI scores, and the low background prevalence of Afib in this cohort (i.e., selection of patients with lower baseline cardiac morbidity) are possible explanations for non-significance of the AI algorithm. As anticipated Age and IPSS-R are the greatest predictors of Death after AlloHCT, although the suggestion of an association with baseline Echo parameters highlights the importance of identifying those who may benefit from preemptive Cardiology intervention.