Farrukh, Faiqa

Introduction: Machine learning (ML) has become increasingly pivotal in health care, particularly in colorectal cancer (CRC) detection and diagnosis with the use of predictive models and artificial intelligence-assisted colonoscopies. This study evaluates the efficacy of ML models in predicting the risk for CRC and advanced colorectal polyps (ACP) before colonoscopy. Methods: A systematic literature review was conducted following PRISMA guidelines, focusing on studies using ML for CRC and ACP prediction. Data extraction regarding study type, ML methodology, quality of data, and bias assessment was in line with the CHARMS checklist. Meta-analysis was also performed to assess the performance of models for the prediction of CRC, adenoma, or both. Results: This systematic review included 14 studies with 3618 median patients (333 to 263,879). Our study demonstrated considerable heterogeneity in methodologies and outcomes, with area under the receiver operating characteristic (AUROC) ranging from 0.6 to 1. The derivation+validation cohorts showed a pooled sensitivity of 0.832 (95% CI: 0.755-0.889) and specificity of 0.802 (95% CI: 0.722-0.863), with an overall AUROC of 0.883. Conclusion: The review underscores the significant role of ML in CRC and ACP diagnosis and its routine use could efficiently direct high-risk patients to timely colonoscopies and spare the low-risk ones from unnecessary procedures. Despite the promise shown, the variability in methodologies and outcomes highlights the need for standardized approaches and further investigation in this field.

Abstract We examined the individual prognostic contribution of absolute neutrophil (ANC), lymphocyte (ALC), and monocyte (AMC) counts, on overall (OS), leukemia‐free (LFS), and myelofibrosis‐free (MFFS) survival in essential thrombocythemia (ET). Informative cases ( N = 598; median age 59 years; females 62%) were retrospectively accrued from a Mayo Clinic database: JAK2 59%, CALR 27%, triple‐negative 11%, and MPL 3%; international prognostic scoring system for ET (IPSET) risk high 21%, intermediate 42%, and low 37%; 7% (37/515) had abnormal karyotype and 10% (21/205) adverse mutations ( SF3B1/SRSF2/U2AF1/TP53 ). At median 8.4 years, 163 (27%) deaths, 71 (12%) fibrotic, and 20 (3%) leukemic transformations were recorded. Multivariable analysis resulted in HR (95% CI) of 16.5 (9.9–27.4) for age > 70 years, 3.7 (2.3–6.0) for age 50–70 years, 2.4 (1.7–3.3) for ANC ≥8 × 10 9 /L, and 1.9 (1.4–2.6) for ALC <1.7 × 10 9 /L. The corresponding HR‐based scores were 4, 2, 1, and 1, resulting in an new 4‐tiered AgeAncAlc (AAA; triple A) risk model: high (5–6 points; median survival 8 years; HR 30.1, 95% CI 17.6–54), intermediate‐2 (4 points; median 13.5 years; HR 12.7, 95% CI 7.1–23.0), intermediate‐1 (2–3 points; median 20.7 years; HR 3.8, 95% CI 2.3–6.4) and low (0–1 points; median 47 years). The AAA model (Akaike Information Criterion [AIC] 621) performed better than IPSET (AIC 647) and was subsequently validated by an external University of Florence ET cohort ( N = 485). None of the AAA variables predicted LFS while ALC <1.7 × 10 9 /L was associated with inferior MFFS ( p = .01). Adverse mutations ( p < .01) and karyotype ( p < .01) displayed additional prognostic value without disqualifying the prognostic integrity of the AAA model. This study proposes a simple and globally applicable survival model for ET, which can be used as a platform for further molecular refinement. This study also suggests a potential role for immune‐related biomarkers, as a prognostic tool in myeloproliferative neoplasms.

AbstractWe retrospectively examined our experience with midostaurin therapy in 33 consecutive patients (median age 68 years; 58% females) with advanced systemic mastocytosis (adv‐SM): aggressive SM (ASM; n = 17), SM associated with another hematologic neoplasm (SM‐AHN; n = 14) and mast cell leukemia (MCL; n = 2). KITD816V mutation was detected in 84% of the patients and C findings in 91%. Eleven (33%) patients were previously treated with other cytoreductive drugs, including cladribine (n = 4) and imatinib (n = 3). Median time from diagnosis to initiation of midostaurin therapy was 2.2 months (range 0.3–41). Using modified valent criteria, overall response was 42% (53% ASM, 29% SM‐AHN, 50% MCL; p = .22), all classified as being major. Responses included ≥50% reduction in bone marrow mast cells in 40% and normalization of serum tryptase in 29%, of evaluated cases. After a median follow‐up of 14.6 months from initiation of midostaurin therapy, 7 (21%) deaths, 1 (3%) leukemic progression, and 18 (55%) treatment discontinuations were documented; median duration of midostaurin treatment was 7.9 months (range 0.5–123) and response duration 21.5 months (range 2.9–123). Most frequent side effect was gastrointestinal (51%) while grade 3/4 neutropenia or thrombocytopenia occurred in 12% of patients. Response to treatment was not predicted by KIT mutation (p = .67) or exposure to prior cytoreductive therapy (p = .44). Median survival was longer in midostaurin responders but not significantly (median 26.5 vs. 16 months; p = .15). Findings from the current study are broadly consistent with previously published clinical trial observations.

Abstract Background The detrimental effect of leukocytosis on survival in myeloproliferative neoplasms (MPN) has been well established for primary myelofibrosis (PMF), polycythemia vera (PV) and essential thrombocythemia (ET ) (JCO. 2018;36:310; BJH. 2020;189:291) Previous studies have also implicated leukocytosis as a risk factor for leukemic transformation (Mayo Clin Proc. 2017;92:1118) and thrombosis in MPN (Blood Adv. 2019;3:1729). However, it is currently not clear as to which component(s) of white blood cells is responsible for these observations. In the current study, we sought to examine the individual prognostic contribution of absolute neutrophil (ANC), lymphocyte (ALC) and monocyte (AMC) counts, on overall (OS), leukemia-free (LFS), and myelofibrosis-free (MFFS) survival and in ET. Methods Study patients (n=349) were retrospectively recruited from the Mayo Clinic MPN database of 1,249 WHO-defined ET patients, evaluated over five decades (1967-2021), based on availability of information on ANC, ALC and AMC. Conventional criteria were used for diagnosis and definitions of major complications, including leukemic or fibrotic transformation (Blood 2016;127:2391). Conventional statistical methods were applied using JMP Pro 14.0.0 software package, SAS Institute, Cary, NC. Multivariable analyses included previously established risk factors for survival. Results 349 patients (median age 57 years, range 18-89; females 61%) with ET were included in the study: 46% JAK2, 34% CALR, 16% triple-negative and 4% MPL mutated; IPSET risk category high 24%, intermediate 41%, and low 35%; presenting median (range) values were 13.8 g/dL (11.1-16.4) for hemoglobin, 8.2 x 10(9)/L (3.2-52) for leukocyte count, and 859 x 10(9)/L (451-3460) for platelet count; palpable splenomegaly was present in 48 (14%); median followup was 10 years (range 0-47), during which time 118 deaths, 52 fibrotic progressions, and 14 leukemic transformations were documented. Multivariable analysis identified older age (p&lt;0.001), increased ANC (p&lt;0.001), decreased ALC (p=0.03), and male sex (p=0.04), but not AMC (p=0.8), venous thrombosis (p=0.4), or arterial thrombosis (p=0.4), as independent risk factors for OS. ANC of ≥8 x 10(9)/L and ALC of &lt;1.8 x 10(9)/L were determined as appropriate cut-off values by ROC analysis. Subsequent multivariable analysis using these cut-off values resulted in HR (95% CI) of 5.2 (3.4-7.9; p&lt;0.001) for age &gt;60 years, 3.1 (2.1-4.6; p&lt;0.001) for ANC, and 2.0 (1.4-3.0; p&lt;0.001) for ALC; male sex was no longer significant in this analysis (p=0.14). An operational HR-based risk score assigned 3 adverse risk points for older age (&gt;60 years), 2 for increased ANC (≥8 x 10(9)/L) and 1 for ALC (&lt;1.8 x 10(9)/L), resulting in an new Age Anc Alc (AAA; triple A) risk model for survival in ET with estimates of median survival ranging from 9.7 to 36.6 years (Figure 1). In addition, ALC &lt;1.8 x 10(9)/L was associated with inferior LFS (p=0.06) and MFFS (p=0.07) while AMC as a continuous variable showed borderline significance for MFFS (p=0.18). In univariate analysis, JAK2V617F allele burden showed significant association with OS (p=0.02), LFS (p=0.05) and MFFS (p=0.001); however significance for OS was lost in mutivariable analysis that included ANC and ALC. An external validation cohort from the University of Florence (n=485) confirmed the independent survival risk contribution from age &gt;60 years (p&lt;0.001; HR 9.9, 95% CI 5.4-18.0), ANC ≥8 x 10(9)/L (p=0.02; 1.9, 1.1-3.5) and ALC &lt;1.8 x 10(9)/L (p&lt;0.001; 2.0, 1.3-3.0). The Triple A risk model was also effectively applied on the Florence validation cohort (Figure 1): median follow-up 8.4 years; 87 deaths; 43 fibrotic progression; 12 leukemic transformations. The association of ALC &lt;1.8 x 10(9)/L (p=0.04) and AMC (p=0.002) with fibrotic transformation was also validated in the Florence cohort. Conclusions: The current study identifies increased ANC and decreased ALC as age-independent risk factors for survival in ET, thus allowing the development of a globally applicable simple to use "Triple A" risk model that is based on Age, ANC and ALC. Decreased ALC also predicted fibrotic and leukemic progression. Our observations suggest potential value for immune profiling as an additional prognostic tool in MPN. Figure 1 Figure 1. Disclosures Szuber: Novartis: Honoraria. Vannucchi: BMS: Honoraria, Membership on an entity's Board of Directors or advisory committees; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees; Incyte: Honoraria, Membership on an entity's Board of Directors or advisory committees; AbbVie: Membership on an entity's Board of Directors or advisory committees.

Abstract Background: Venetoclax, a small-molecule inhibitor of B cell leukemia/lymphoma-2, in combination with hypomethylating agents (HMA) has shown improved efficacy and survival benefit compared to HMA alone (DiNardo et al, 2020) in elderly/unfit patients with acute myeloid leukemia (AML). Since FDA-approval of this regimen for elderly/unfit AML patients, it is frequently utilized both in the upfront and relapsed/refractory setting. Cardiac events with venetoclax are not well described. In the VIALE-E trial, which included patients ineligible for standard induction chemotherapy due to congestive heart failure or stable angina, 15% of patients receiving azacitidine plus venetoclax experienced atrial fibrillation as a serious adverse event, vs. 1% in the azacitidine plus placebo group (DiNardo et al, 2020). Our objective was to provide an estimate of the prevalence of and a description of all cardiac events that occurred in AML patients undergoing treatment with venetoclax + HMA. Methods: 170 consecutive patients with AML who received venetoclax +HMA (azacitidine or decitabine) outside the context of a clinical trial between 1/2017-11/2020 at the Mayo Clinic were included. Patients received venetoclax + HMA either as upfront treatment or for relapsed/refractory disease. Patients with relapse following allogeneic stem cell transplant were excluded. We evaluated all cardiac events that occurred while treatment with venetoclax + HMA was ongoing. Baseline patient and treatment characteristics were compared using the Mann-Whitney U-test and the Fisher's exact test. All statistics were computed using EZR (Version 1.53). Results: 1. Patient characteristics A total of 170 patients who received venetoclax + HMA (median age 69 years [range 17-91], 63% males) were included. ELN risk category was either adverse risk (48%, 82/170) or intermediate risk (48%, 82/170) in the majority of patients. 64% (109/170) of patients received venetoclax + HMA as upfront treatment. Characteristics including age, ELN cytogenetic risk, cardiovascular risk factors, and upfront vs relapsed therapy were similar among patients with or without cardiac events. The only exception was a higher incidence of CEBPA mutation amongst those with cardiac events (12% vs 2%, p=0.03). The majority (83%, 141/170) of patients underwent an echocardiogram prior to initiation of therapy. 2. Cardiac events Of 170 patients treated with venetoclax + HMA during the study period, 34 (20%) patients experienced a total of 48 cardiac events. Of patients experiencing cardiac events, 32% (11/34) had no pre-existing cardiac disease and 12% (4/34) had no cardiovascular risk factors (Table 1). The majority of events occurred early in treatment course: 41% during cycle 1, 26% during cycle 2 and 15% during cycle 3 (Table 1). The most frequently occurring cardiac event (21%, 10/48 events) was a decrease in left ventricular ejection fraction on echocardiography, which was associated with symptoms in all ten patients. Second most frequent was atrial fibrillation with rapid ventricular response at 17% (8/48 events), followed by troponin elevation without electrocardiogram changes at 15% (7/46 events). Of patients with troponin elevation, 57% (4/7 events) occurred in the setting of another inciting factor such as severe anemia, while 43% represented a troponin elevation without explanation (Table 1). Other cardiac events included heart failure with preserved ejection fraction (n=4), other symptomatic arrhythmia (n=4), and symptomatic pericardial effusion or pericarditis (n=3). In addition, 2 of 34 (6%) patients experienced fatal cardiopulmonary arrest. The majority (88%) of cardiac events required either inpatient admission (62%, 21/34 patients) or intensive care unit (ICU) care (26%, 9/34 patients). 77% of patients required new cardiac medications or procedural intervention (n=4). In 27% of cases (9/34 patients), the cardiac event directly contributed to death (Table 1). Conclusions: Cardiac complications were observed in one-fifth (20%) of AML patients treated with venetoclax + HMA, despite the absence of preexisting cardiac disease in a third of cases; moreover 27% of events were fatal. Further comparative studies are required to identify salient clinical features predictive of cardiac complications in these patients. Figure 1 Figure 1. Disclosures Al-Kali: Novartis: Research Funding; Astex: Other: Research support to institution. Litzow: Astellas: Research Funding; Amgen: Research Funding; Omeros: Other: Advisory Board; AbbVie: Research Funding; Actinium: Research Funding; Pluristem: Research Funding; Jazz: Other: Advisory Board; Biosight: Other: Data monitoring committee. Patnaik: Kura Oncology: Research Funding; StemLine: Research Funding.