Endpoint Surrogacy in First-Line Chronic Lymphocytic Leukemia.

PURPOSE: Surrogate endpoints are commonly used to estimate treatment efficacy in clinical studies of chronic lymphocytic leukemia (CLL). This patient- and trial-level analysis describes the correlation between progression-free survival (PFS) and minimal residual disease (MRD) with overall survival (OS) in first-line trials for CLL. PATIENTS AND METHODS: First, patient-level correlation was confirmed using source data from 12 front-line GCLLSG-trials. Additionally, a joint-frailty copula model was fitted to validate correlation in the setting of targeted therapies.Second, a meta-analysis of first-line phase III trials in CLL from 2008-2024 was performed. Treatment effect correlation was quantified from seven GCLLSG and nine published trials, using hazard ratios for time-to-event and odds ratios for binary endpoints. RESULTS: The GCLLSG analysis set comprised 4237 patients. Patient-level correlation for PFS/OS was strong with Spearman's Rho >0.9. The joint-frailty copula indicated a weak correlation for C/CIT with a tau of 0.52, (95% CI: 0.49 - 0.55) while the correlation was strong for TT (tau = 0.91, 95% CI: 0.89 - 0.93).The meta-analysis set contained a total of 8065 patients including 5198 (64%) patients treated with C/CIT and 2867 (36%) treated with TT. Treatment effect correlation of the hazard ratios (HR) for PFS and OS was R = 0.75 (95% CI: 0.74 - 0.76, R2 = 0.56), while correlation of end-of-treatment MRD with PFS and OS was R = 0.88 (95%CI: -0.87 - 0.89; R2 = 0.78) and 0.71 (95%CI: 0.69 - 0.73; R2 = 0.5), respectively. CONCLUSION: Patient-level correlation was confirmed in the setting of targeted therapies while treatment-effect correlation between PFS and OS remains uncertain. MRD response status showed a high treatment-effect correlation with PFS but not OS, with the caveat of a limited number of randomized trials with available MRD data.

Venetoclax-Obinutuzumab for previously untreated chronic lymphocytic leukemia: 6-year results of the phase 3 CLL14 study.

Venetoclax-obinutuzumab (Ven-Obi) is a standard-of-care for patients with previously untreated chronic lymphocytic leukemia (CLL). In the CLL14 study, patients with previously untreated CLL and coexisting conditions were randomized to 12 cycles of Ven-Obi (n=216) or chlorambucil-obinutuzumab (Clb-Obi, n=216). Progression-free survival (PFS) was the primary endpoint. Key secondary endpoints included time-to-next-treatment (TTNT), rates of undetectable minimal residual disease (uMRD), overall survival (OS) and rates of adverse events. Patient reported outcomes (PROs) of time until definitive deterioration (TUDD) in quality of life (QoL) were analyzed. At a median observation time of 76.4 months, PFS remained superior for Ven-Obi compared to Clb-Obi (median 76.2 vs 36.4 months; HR 0.40[95%CI 0.31-0.52], p<0.0001). Likewise, TTNT was longer after Ven-Obi (6-year-TTNT 65.2% vs 37.1%; HR 0.44, 95%CI 0.33-0.58, p<0.0001). In the Ven-Obi arm, presence of del(17p), unmutated-IGHV and lymph node size ≥5 cm were independent prognostic factors for shorter PFS. Five years after treatment, 17 patients (7.9% of intention-to-treat-population) in the Ven-Obi arm had uMRD (<10-4 in peripheral blood) compared to 4 (1.9%) in the Clb-Obi arm. 6-year-OS rate was 78.7% in the Ven-Obi and 69.2% in the Clb-Obi arm (HR 0.69[95%CI 0.48-1.01], p=0.052). A significantly longer TUDD in global health status/QoL was observed in the Ven-Obi compared to the Clb-Obi arm (median 82.1 vs 65.1 months; HR 0.70[95%CI 0.51-0.97]). Follow-up adjusted SPM incidence rates were 2.3 and 1.4/1000 patient-months in the Ven-Obi and Clb-Obi arm, respectively. The sustained long-term survival, uMRD and QoL benefits support the use of one-year fixed-duration Ven-Obi in CLL. NCT02242942, EudraCT 2014-001810-24.

A retrieval-augmented chatbot based on GPT-4 provides appropriate differential diagnosis in gastrointestinal radiology: a proof of concept study.

BACKGROUND: We investigated the potential of an imaging-aware GPT-4-based chatbot in providing diagnoses based on imaging descriptions of abdominal pathologies. METHODS: Utilizing zero-shot learning via the LlamaIndex framework, GPT-4 was enhanced using the 96 documents from the Radiographics Top 10 Reading List on gastrointestinal imaging, creating a gastrointestinal imaging-aware chatbot (GIA-CB). To assess its diagnostic capability, 50 cases on a variety of abdominal pathologies were created, comprising radiological findings in fluoroscopy, MRI, and CT. We compared the GIA-CB to the generic GPT-4 chatbot (g-CB) in providing the primary and 2 additional differential diagnoses, using interpretations from senior-level radiologists as ground truth. The trustworthiness of the GIA-CB was evaluated by investigating the source documents as provided by the knowledge-retrieval mechanism. Mann-Whitney U test was employed. RESULTS: The GIA-CB demonstrated a high capability to identify the most appropriate differential diagnosis in 39/50 cases (78%), significantly surpassing the g-CB in 27/50 cases (54%) (p = 0.006). Notably, the GIA-CB offered the primary differential in the top 3 differential diagnoses in 45/50 cases (90%) versus g-CB with 37/50 cases (74%) (p = 0.022) and always with appropriate explanations. The median response time was 29.8 s for GIA-CB and 15.7 s for g-CB, and the mean cost per case was $0.15 and $0.02, respectively. CONCLUSIONS: The GIA-CB not only provided an accurate diagnosis for gastrointestinal pathologies, but also direct access to source documents, providing insight into the decision-making process, a step towards trustworthy and explainable AI. Integrating context-specific data into AI models can support evidence-based clinical decision-making. RELEVANCE STATEMENT: A context-aware GPT-4 chatbot demonstrates high accuracy in providing differential diagnoses based on imaging descriptions, surpassing the generic GPT-4. It provided formulated rationale and source excerpts supporting the diagnoses, thus enhancing trustworthy decision-support. KEY POINTS: • Knowledge retrieval enhances differential diagnoses in a gastrointestinal imaging-aware chatbot (GIA-CB). • GIA-CB outperformed the generic counterpart, providing formulated rationale and source excerpts. • GIA-CB has the potential to pave the way for AI-assisted decision support systems.

First-line venetoclax combinations versus chemoimmunotherapy in fit patients with chronic lymphocytic leukaemia (GAIA/CLL13): 4-year follow-up from a multicentre, open-label, randomised, phase 3 trial.

BACKGROUND: In the primary analysis report of the GAIA/CLL13 trial, we found that venetoclax-obinutuzumab and venetoclax-obinutuzumab-ibrutinib improved undetectable measurable residual disease (MRD) rates and progression-free survival compared with chemoimmunotherapy in patients with previously untreated chronic lymphocytic leukaemia. However, to our knowledge, no data on direct comparisons of different venetoclax-based combinations are available. METHODS: GAIA/CLL13 is an open-label, randomised, phase 3 study conducted at 159 sites in ten countries in Europe and the Middle East. Eligible patients were aged 18 years or older, with a life expectancy of at least 6 months, an Eastern Cooperative Oncology group performance status of 0-2, a cumulative illness rating scale score of 6 or lower or a single score of 4 or lower, and no TP53 aberrations. Patients were randomly assigned (1:1:1:1), with a computer-generated list stratified by age, Binet stage, and regional study group, to either chemoimmunotherapy, venetoclax-rituximab, venetoclax-obinutuzumab, or venetoclax-obinutuzumab-ibrutinib. All treatments were administered in 28-day cycles. Patients in the chemoimmunotherapy group received six cycles of treatment, with patients older than 65 years receiving intravenous bendamustine (90 mg/m2, days 1-2), whereas patients aged 65 years or younger received intravenous fludarabine (25 mg/m2, days 1-3) and intravenous cyclophosphamide (250 mg/m2, days 1-3). Intravenous rituximab (375 mg/m2, day 1 of cycle 1; 500 mg/m2, day 1 of cycles 2-6) was added to chemotherapy. In the experimental groups, patients received daily venetoclax (400 mg orally) for ten cycles after a 5-week ramp-up phase starting on day 22 of cycle 1. In the venetoclax-rituximab group, intravenous rituximab (375 mg/m2, day 1 of cycle 1; 500 mg/m2, day 1 of cycles 2-6) was added. In the obinutuzumab-containing groups, obinutuzumab was added (cycle 1: 100 mg on day 1, 900 mg on day 2, and 1000 mg on days 8 and 15; cycles 2-6: 1000 mg on day 1). In the venetoclax-obinutuzumab-ibrutinib group, daily ibrutinib (420 mg orally, from day 1 of cycle 1) was added until undetectable MRD was reached in two consecutive measurements (3 months apart) or until cycle 36. The planned treatment duration was six cycles in the chemoimmunotherapy group, 12 cycles in the venetoclax-rituximab and the venetoclax-obinutuzumab group and between 12 and 36 cycles in the venetoclax-obinutuzumab-ibrutinib group. Coprimary endpoints were the undetectable MRD rate in peripheral blood at month 15 for the comparison of venetoclax-obinutuzumab versus standard chemoimmunotherapy and investigator-assessed progression-free survival for the comparison of venetoclax-obinutuzumab-ibrutinib versus standard chemoimmunotherapy, both analysed in the intention-to-treat population (ie, all patients randomly assigned to treatment) with a split α of 0·025 for each coprimary endpoint. Both coprimary endpoints have been reported elsewhere. Here we report a post-hoc exploratory analysis of updated progression-free survival results after a 4-year follow-up of our study population. Safety analyses included all patients who received at least one dose of study treatment. This study is registered with ClinicalTrials.gov, NCT02950051, recruitment is complete, and all patients are off study treatment. FINDINGS: Between Dec 13, 2016, and Oct 13, 2019, 1080 patients were screened and 926 were randomly assigned to treatment (chemoimmunotherapy group n=229; venetoclax-rituximab group n=237; venetoclax-obinutuzumab group n=229; and venetoclax-obinutuzumab-ibrutinib group n=231); mean age 60·8 years (SD 10·2), 259 (28%) of 926 patients were female, and 667 (72%) were male (data on race and ethnicity are not reported). At data cutoff for this exploratory follow-up analysis (Jan 31, 2023; median follow-up 50·7 months [IQR 44·6-57·9]), patients in the venetoclax-obinutuzumab group had significantly longer progression-free survival than those in the chemoimmunotherapy group (hazard ratio [HR] 0·47 [97·5% CI 0·32-0·69], p<0·0001) and the venetoclax-rituximab group (0·57 [0·38-0·84], p=0·0011). The venetoclax-obinutuzumab-ibrutinib group also had a significantly longer progression-free survival than the chemoimmunotherapy group (0·30 [0·19-0·47]; p<0·0001) and the venetoclax-rituximab group (0·38 [0·24-0·59]; p<0·0001). There was no difference in progression-free survival between the venetoclax-obinutuzumab-ibrutinib and venetoclax-obinutuzumab groups (0·63 [0·39-1·02]; p=0·031), and the proportional hazards assumption was not met for the comparison between the venetoclax-rituximab group versus the chemoimmunotherapy group (log-rank p=0·10). The estimated 4-year progression-free survival rate was 85·5% (97·5% CI 79·9-91·1; 37 [16%] events) in the venetoclax-obinutuzumab-ibrutinib group, 81·8% (75·8-87·8; 55 [24%] events) in the venetoclax-obinutuzumab group, 70·1% (63·0-77·3; 84 [35%] events) in the venetoclax-rituximab group, and 62·0% (54·4-69·7; 90 [39%] events) in the chemoimmunotherapy group. The most common grade 3 or worse treatment-related adverse event was neutropenia (114 [53%] of 216 patients in the chemoimmunotherapy group, 109 [46%] of 237 in the venetoclax-rituximab group, 127 [56%] of 228 in the venetoclax-obinutuzumab group, and 112 [48%] of 231 in the venetoclax-obinutuzumab-ibrutinib group). Deaths determined to be associated with study treatment by the investigator occurred in three (1%) patients in the chemoimmunotherapy group (n=1 due to each of sepsis, metastatic squamous cell carcinoma, and Richter's syndrome), none in the venetoclax-rituximab and venetoclax-obinutuzumab groups, and four (2%) in the venetoclax-obinutuzumab-ibrutinib group (n=1 due to each of acute myeloid leukaemia, fungal encephalitis, small-cell lung cancer, and toxic leukoencephalopathy). INTERPRETATION: With more than 4 years of follow-up, venetoclax-obinutuzumab and venetoclax-obinutuzumab-ibrutinib significantly extended progression-free survival compared with both chemoimmunotherapy and venetoclax-rituximab in previously untreated, fit patients with chronic lymphocytic leukaemia, thereby supporting their use and further evaluation in this patient group, while still considering the higher toxicities observed with the triple combination. FUNDING: AbbVie, Janssen, and F Hoffmann-La Roche.

Fludarabine, cyclophosphamide, and rituximab as first-line treatment in patients with chronic lymphocytic leukemia: A long-term analysis of the German CLL Study Group (GCLLSG) registry.

Long-term data of chronic lymphocytic leukemia (CLL) patients with favorable risk who were treated with fludarabine, cyclophosphamide, and rituximab (FCR) within clinical trials show good efficacy. We here report long-term data collected within the GCLLSG registry. Altogether, 417 CLL patients who received first-line treatment with FCR were analyzed, of which 293 (70.3%) were treated outside of clinical trials. The median observation time from first-line was 95.8 (interquartile range 58.7-126.8) months. Focusing on data of 194 (46.5%) patients who received FCR first-line treatment after 2013 (start of data collection within GCLLSG registry), responses were documented in 85% of the patients, non-responses in 15%, and for 3.6% the assessment was missing. Median event-free survival (EFS, time until disease progression, subsequent treatment, or death) was 60.2 months with a 5-year EFS-rate of 50.6%. Patients with higher-risk disease, characterized by unmutated IGHV (N = 78), had a median EFS of 45.4 months with a 5-year EFS rate of 36.3%, while the median EFS was 77.5 months with a 5-year EFS rate of 60.3% in patients with mutated IGHV (N = 40). Median overall survival was not reached with a 5-year survival rate of 92.7%. In summary, first-line FCR was associated with long EFS, especially in patients exhibiting a mutated IGHV status.

Reassessing the chronic lymphocytic leukemia International Prognostic Index in the era of targeted therapies.

ABSTRACT: We evaluated the chronic lymphocytic leukemia International Prognostic Index (CLL-IPI) in patients with CLL treated first line with targeted drugs (n = 991) or chemoimmunotherapy (n = 1256). With a median observation time of 40.5 months, the 3-year progression-free survival (PFS) rates for targeted drug-treated patients varied by CLL-IPI risk group: 96.5% (low), 87.6% (intermediate), 82.4% (high), and 78.7% (very high). Differences between consecutive CLL-IPI risk groups were observed for intermediate vs low and high vs intermediate, but not very high vs high. CLL-IPI factors ß2-microglobulin, immunoglobulin heavy variable (IGHV) status, and TP53 status each retained prognostic value for PFS. The 3-year overall survival (OS) rates by CLL-IPI risk groups were 100%, 96%, 93.9%, and 89.4%, respectively, with no differences between consecutive risk groups. Age, Binet stage, ß2-microglobulin, and TP53 status each retained prognostic value for OS. In chemoimmunotherapy patients (median observation time, 66.9 months), 3-year PFS rates for CLL-IPI risk groups were 78.1%, 51.4%, 40.1%, and 16.5%, respectively; corresponding 3-year OS rates were 97.4%, 93.1%, 81.8%, and 57.3%. In a matched-pair analysis, PFS differences in targeted therapies (n = 812) vs chemoimmunotherapy (n = 812) across all risk groups and OS differences in all but patients at low risk were demonstrated. The CLL-IPI maintains its prognostic value in predicting PFS outcomes with targeted drugs, but its impact in predicting survival appears diminished. Targeted therapies showed enhanced outcomes over chemoimmunotherapy, highlighting their effectiveness across various risk groups. Our findings support ongoing assessment of prognostic tools in CLL treatment evolution. These trials were registered at www.ClinicalTrials.gov as #NCT02345863, #NCT02401503, #NCT02689141, #NCT02445131, #NCT02758665, #NCT02950051, #NCT02242942, #NCT00262782, #NCT00281918, and #NCT01010061.

Acalabrutinib, venetoclax, and obinutuzumab in relapsed/refractory CLL: final efficacy and ctDNA analysis of the CLL2-BAAG trial.

ABSTRACT: The phase 2 CLL2-BAAG trial tested the measurable residual disease (MRD)-guided triple combination of acalabrutinib, venetoclax, and obinutuzumab after optional bendamustine debulking in 45 patients with relapsed/refractory chronic lymphocytic leukemia (CLL). MRD was measured by flow cytometry (FCM; undetectable MRD <10-4) in peripheral blood (PB) and circulating tumor DNA (ctDNA) using digital droplet polymerase chain reaction of variable-diversity-joining (VDJ) rearrangements and CLL-related mutations in plasma. The median number of previous treatments was 1 (range, 1-4); 18 patients (40%) had received a Bruton tyrosine kinase inhibitor (BTKi) and/or venetoclax before inclusion, 14 of 44 (31.8%) had TP53 aberrations, and 34 (75.6%) had unmutated immunoglobulin heavy-chain variable region genes. With a median observation time of 36.3 months and all patients off-treatment for a median of 21.9 months, uMRD <10-4 in PB was achieved in 42 of the 45 patients (93.3%) at any time point, including 17 of 18 (94.4%) previously exposed to venetoclax/BTKi and 13 of 14 (92.9%) with TP53 aberrations. The estimated 3-year progression-free and overall survival rates were 85.0% and 93.8%, respectively. Overall, 585 paired FCM/ctDNA samples were analyzed and 18 MRD recurrences (5 with and 13 without clinical progression) occurred after the end of treatment. Twelve samples were first detected by ctDNA, 3 by FCM, and 3 synchronously. In conclusion, time-limited MRD-guided acalabrutinib, venetoclax, and obinutuzumab achieved deep remissions in almost all patients with relapsed/refractory CLL. The addition of ctDNA-based analyses to FCM MRD assessment seems to improve early detection of relapses. This trial was registered at www.clinicaltrials.gov as #NCT03787264.

A deep learning approach for projection and body-side classification in musculoskeletal radiographs.

BACKGROUND: The growing prevalence of musculoskeletal diseases increases radiologic workload, highlighting the need for optimized workflow management and automated metadata classification systems. We developed a large-scale, well-characterized dataset of musculoskeletal radiographs and trained deep learning neural networks to classify radiographic projection and body side. METHODS: In this IRB-approved retrospective single-center study, a dataset of musculoskeletal radiographs from 2011 to 2019 was retrieved and manually labeled for one of 45 possible radiographic projections and the depicted body side. Two classification networks were trained for the respective tasks using the Xception architecture with a custom network top and pretrained weights. Performance was evaluated on a hold-out test sample, and gradient-weighted class activation mapping (Grad-CAM) heatmaps were computed to visualize the influential image regions for network predictions. RESULTS: A total of 13,098 studies comprising 23,663 radiographs were included with a patient-level dataset split, resulting in 19,183 training, 2,145 validation, and 2,335 test images. Focusing on paired body regions, training for side detection included 16,319 radiographs (13,284 training, 1,443 validation, and 1,592 test images). The models achieved an overall accuracy of 0.975 for projection and 0.976 for body-side classification on the respective hold-out test sample. Errors were primarily observed in projections with seamless anatomical transitions or non-orthograde adjustment techniques. CONCLUSIONS: The deep learning neural networks demonstrated excellent performance in classifying radiographic projection and body side across a wide range of musculoskeletal radiographs. These networks have the potential to serve as presorting algorithms, optimizing radiologic workflow and enhancing patient care. RELEVANCE STATEMENT: The developed networks excel at classifying musculoskeletal radiographs, providing valuable tools for research data extraction, standardized image sorting, and minimizing misclassifications in artificial intelligence systems, ultimately enhancing radiology workflow efficiency and patient care. KEY POINTS: • A large-scale, well-characterized dataset was developed, covering a broad spectrum of musculoskeletal radiographs. • Deep learning neural networks achieved high accuracy in classifying radiographic projection and body side. • Grad-CAM heatmaps provided insight into network decisions, contributing to their interpretability and trustworthiness. • The trained models can help optimize radiologic workflow and manage large amounts of data.

Hybrid immunity to SARS-CoV-2 in patients with chronic lymphocytic leukemia.

OBJECTIVE: Preventing severe COVID-19 remains a priority globally, particularly in the immunocompromised population. As shown in healthy individuals, immunity against SARS-CoV-2 can be yielded by previous infection, vaccination, or both (hybrid immunity). The objective of this observation study was to investigate hybrid immunity in patients with chronic lymphocytic leukemia (CLL). METHODS/RESULTS: Blood samples of six patients with CLL were collected 55 days after fourth COVID-19 vaccination. All patients had a SARS-CoV-2 infection within 12 months before the second booster (fourth vaccination). SARS-CoV-2 spike receptor binding domain (RBD)-specific IgG antibodies were detectable in 6/6 (100.0%) CLL patients after four compared to 4/6 (66.7%) after three vaccinations. The median number of SARS-CoV-2 spike-specific T cells after repeated booster vaccination plus infection was 166 spot-forming cells (SFC) per million peripheral blood mononuclear cells. Overall, 5/5 (100%) studied patients showed a detectable increase in T cell activity. CONCLUSION: Our data reveal an increase of cellular and humoral immune response in CLL patients after fourth COVID-19 vaccination combined with SARS-CoV-2 infection, even in those undergoing B cell-depleting treatment. Patients with prior vaccination failure now show a specific IgG response. Future research should explore the duration and effectiveness of hybrid immunity considering various factors like past infection and vaccination rates, types and numbers of doses, and emerging variants.

Tislelizumab plus zanubrutinib for Richter transformation: the phase 2 RT1 trial.

In patients with chronic lymphocytic leukemia, Richter transformation (RT) reflects the development of an aggressive lymphoma that is associated with poor response to chemotherapy and short survival. We initiated an international, investigator-initiated, prospective, open-label phase 2 study in which patients with RT received a combination of the PD-1 inhibitor tislelizumab plus the BTK inhibitor zanubrutinib for 12 cycles. Patients responding to treatment underwent maintenance treatment with both agents. The primary end point was overall response rate after six cycles. Of 59 enrolled patients, 48 patients received at least two cycles of treatment and comprised the analysis population according to the study protocol. The median observation time was 13.9 months, the median age was 67 (range 45-82) years. Ten patients (20.8%) had received previous RT-directed therapy. In total, 28 out of 48 patients responded to induction therapy with an overall response rate of 58.3% (95% confidence interval (CI) 43.2-72.4), including 9 (18.8%) complete reponse and 19 (39.6%) partial response, meeting the study's primary end point by rejecting the predefined null hypothesis of 40% (P = 0.008). Secondary end points included duration of response, progression-free survival and overall survival. The median duration of response was not reached, the median progression-free survival was 10.0 months (95% CI 3.8-16.3). Median overall survival was not reached with a 12-month overall survival rate of 74.7% (95% CI 58.4-91.0). The most common adverse events were infections (18.0%), gastrointestinal disorders (13.0%) and hematological toxicities (11.4%). These data suggest that combined checkpoint and BTK inhibition by tislelizumab plus zanubrutinib is an effective and well-tolerated treatment strategy for patients with RT. ClinicalTrials.gov Identifier: NCT04271956 .

Performance of ChatGPT, human radiologists, and context-aware ChatGPT in identifying AO codes from radiology reports.

While radiologists can describe a fracture's morphology and complexity with ease, the translation into classification systems such as the Arbeitsgemeinschaft Osteosynthesefragen (AO) Fracture and Dislocation Classification Compendium is more challenging. We tested the performance of generic chatbots and chatbots aware of specific knowledge of the AO classification provided by a vector-index and compared it to human readers. In the 100 radiological reports we created based on random AO codes, chatbots provided AO codes significantly faster than humans (mean 3.2 s per case vs. 50 s per case, p < .001) though not reaching human performance (max. chatbot performance of 86% correct full AO codes vs. 95% in human readers). In general, chatbots based on GPT 4 outperformed the ones based on GPT 3.5-Turbo. Further, we found that providing specific knowledge substantially enhances the chatbot's performance and consistency as the context-aware chatbot based on GPT 4 provided 71% consistent correct full AO codes for the compared to the 2% consistent correct full AO codes for the generic ChatGPT 4. This provides evidence, that refining and providing specific context to ChatGPT will be the next essential step in harnessing its power.

A Context-based Chatbot Surpasses Trained Radiologists and Generic ChatGPT in Following the ACR Appropriateness Guidelines.

Background Radiological imaging guidelines are crucial for accurate diagnosis and optimal patient care as they result in standardized decisions and thus reduce inappropriate imaging studies. Purpose In the present study, we investigated the potential to support clinical decision-making using an interactive chatbot designed to provide personalized imaging recommendations from American College of Radiology (ACR) appropriateness criteria documents using semantic similarity processing. Methods We utilized 209 ACR appropriateness criteria documents as specialized knowledge base and employed LlamaIndex, a framework that allows to connect large language models with external data, and the ChatGPT 3.5-Turbo to create an appropriateness criteria contexted chatbot (accGPT). Fifty clinical case files were used to compare the accGPT's performance against general radiologists at varying experience levels and to generic ChatGPT 3.5 and 4.0. Results All chatbots reached at least human performance level. For the 50 case files, the accGPT performed best in providing correct recommendations that were "usually appropriate" according to the ACR criteria and also did provide the highest proportion of consistently correct answers in comparison with generic chatbots and radiologists. Further, the chatbots provided substantial time and cost savings, with an average decision time of 5 minutes and a cost of 0.19 € for all cases, compared to 50 minutes and 29.99 € for radiologists (both p < 0.01). Conclusion ChatGPT-based algorithms have the potential to substantially improve the decision-making for clinical imaging studies in accordance with ACR guidelines. Specifically, a context-based algorithm performed superior to its generic counterpart, demonstrating the value of tailoring AI solutions to specific healthcare applications.

Treatment with idelalisib in patients with chronic lymphocytic leukemia - real world data from the registry of the German CLL Study Group.

Idelalisib in combination with rituximab is an efficacious treatment for patients suffering from chronic lymphocytic leukemia (CLL) with known limitations due to toxicities. However, the benefit after prior Bruton tyrosine kinase inhibitor (BTKi) therapy remains unclear. For this analysis, 81 patients included in a non-interventional registry study of the German CLL study group (registered at www.clinicaltrials.gov as # NCT02863692) meeting the predefined criteria of a confirmed diagnosis of CLL and being treated with idelalisib containing regimens outside clinical trials were considered. 11 patients were treatment naïve (13.6%) and 70 patients (86.4%) pretreated. Patients had median of one prior therapy line (range 0-11). Median treatment duration with idelalisib was 5.1 months (range 0-55.0 months). Of 58 patients with documented treatment outcome, 39 responded to idelalisib containing therapy (67.2%). Patients treated with the BTKi ibrutinib as last prior treatment prior to idelalisib responded in 71.4% compared to a response rate of 61.9% in patients without prior ibrutinib. Median event free survival (EFS) was 15.9 months with a 16 versus 14 months EFS in patients with ibrutinib as last prior treatment or not, respectively. Median overall survival was 46.6 months. In conclusion, treatment with idelalisib appears to have a valuable impact in patients being refractory to prior ibrutinib therapy even though there are limitations in our analysis due to the low number of patients included.

Final analysis of the CLL2-GIVe trial: obinutuzumab, ibrutinib, and venetoclax for untreated CLL with del(17p)/TP53mut.

The final analysis of the open-label, multicenter phase 2 CLL2-GIVe trial shows response and tolerability of the triple combination of obinutuzumab, ibrutinib, and venetoclax (GIVe regimen) in 41 previously untreated patients with high-risk chronic lymphocytic leukemia (CLL) with del(17p) and/or TP53 mutation. Induction consisted of 6 cycles of GIVe; venetoclax and ibrutinib were continued up to cycle 12 as consolidation. Ibrutinib was given until cycle 15 or up to cycle 36 in patients not achieving a complete response and with detectable minimal residual disease. The primary end point was the complete remission rate at cycle 15, which was achieved at 58.5% (95% CI, 42.1-73.7; P < .001). The last patient reached the end of the study in January 2022. After a median observation time of 38.4 months (range, 3.7-44.9), the 36-month progression-free survival was 79.9%, and the 36-month overall survival was 92.6%. Only 6 patients continued ibrutinib maintenance. Adverse events of concern were neutropenia (48.8%, grade ≥3) and infections (19.5%, grade ≥3). Cardiovascular toxicity grade 3 occurred as atrial fibrillation at a rate of 2.4% between cycles 1 and 12, as well as hypertension (4.9%) between cycles 1 and 6. The incidence of adverse events of any grade and grade ≥3 was highest during induction and decreased over time. Progressive disease was observed in 7 patients between cycles 27 and 42. In conclusion, the CLL2-GIVe regimen is a promising fixed-duration, first-line treatment for patients with high-risk CLL with a manageable safety profile.

First-Line Venetoclax Combinations in Chronic Lymphocytic Leukemia.

BACKGROUND: Randomized trials of venetoclax plus anti-CD20 antibodies as first-line treatment in fit patients (i.e., those with a low burden of coexisting conditions) with advanced chronic lymphocytic leukemia (CLL) have been lacking. METHODS: In a phase 3, open-label trial, we randomly assigned, in a 1:1:1:1 ratio, fit patients with CLL who did not have TP53 aberrations to receive six cycles of chemoimmunotherapy (fludarabine-cyclophosphamide-rituximab or bendamustine-rituximab) or 12 cycles of venetoclax-rituximab, venetoclax-obinutuzumab, or venetoclax-obinutuzumab-ibrutinib. Ibrutinib was discontinued after two consecutive measurements of undetectable minimal residual disease or could be extended. The primary end points were undetectable minimal residual disease (sensitivity, <10-4 [i.e., <1 CLL cell in 10,000 leukocytes]) as assessed by flow cytometry in peripheral blood at month 15 and progression-free survival. RESULTS: A total of 926 patients were assigned to one of the four treatment regimens (229 to chemoimmunotherapy, 237 to venetoclax-rituximab, 229 to venetoclax-obinutuzumab, and 231 to venetoclax-obinutuzumab-ibrutinib). At month 15, the percentage of patients with undetectable minimal residual disease was significantly higher in the venetoclax-obinutuzumab group (86.5%; 97.5% confidence interval [CI], 80.6 to 91.1) and the venetoclax-obinutuzumab-ibrutinib group (92.2%; 97.5% CI, 87.3 to 95.7) than in the chemoimmunotherapy group (52.0%; 97.5% CI, 44.4 to 59.5; P<0.001 for both comparisons), but it was not significantly higher in the venetoclax-rituximab group (57.0%; 97.5% CI, 49.5 to 64.2; P = 0.32). Three-year progression-free survival was 90.5% in the venetoclax-obinutuzumab-ibrutinib group and 75.5% in the chemoimmunotherapy group (hazard ratio for disease progression or death, 0.32; 97.5% CI, 0.19 to 0.54; P<0.001). Progression-free survival at 3 years was also higher with venetoclax-obinutuzumab (87.7%; hazard ratio for disease progression or death, 0.42; 97.5% CI, 0.26 to 0.68; P<0.001), but not with venetoclax-rituximab (80.8%; hazard ratio, 0.79; 97.5% CI, 0.53 to 1.18; P = 0.18). Grade 3 and grade 4 infections were more common with chemoimmunotherapy (18.5%) and venetoclax-obinutuzumab-ibrutinib (21.2%) than with venetoclax-rituximab (10.5%) or venetoclax-obinutuzumab (13.2%). CONCLUSIONS: Venetoclax-obinutuzumab with or without ibrutinib was superior to chemoimmunotherapy as first-line treatment in fit patients with CLL. (Funded by AbbVie and others; GAIA-CLL13 ClinicalTrials.gov number, NCT02950051; EudraCT number, 2015-004936-36.).