The treatment landscape for multiple myeloma (MM) has experienced substantial progress over the last decade. Despite the efficacy of new substances, patient responses tend to still be highly unpredictable. With increasing cognitive burden that is introduced through a complex and evolving treatment landscape, data-driven assistance tools are becoming more and more popular. Model-based approaches, such as digital twins (DT), enable simulation of probable responses to a set of input parameters based on retrospective observations. In the context of treatment decision-support, those mechanisms serve the goal to predict therapeutic outcomes to distinguish a favorable option from a potential failure. In the present work, we propose a similarity-based multiple myeloma digital twin (MMDT) that emphasizes explainability and interpretability in treatment outcome evaluation. We've conducted a requirement specification process using scientific literature from the medical and methodological domains to derive an architectural blueprint for the design and implementation of the MMDT. In a subsequent stage, we've implemented a four-layer concept where for each layer, we describe the utilized implementation procedure and interfaces to the surrounding DT environment. We further specify our solutions regarding the adoption of multi-line treatment strategies, the integration of external evidence and knowledge, as well as mechanisms to enable transparency in the data processing logic. Furthermore, we define an initial evaluation scenario in the context of patient characterization and treatment outcome simulation as an exemplary use case for our MMDT. Our derived MMDT instance is defined by 475 unique entities connected through 438 edges to form a MM knowledge graph. Using the MMRF CoMMpass real-world evidence database and a sample MM case, we processed a complete outcome assessment. The output shows a valid selection of potential treatment strategies for the integrated medical case and highlights the potential of the MMDT to be used for such applications. DT models face significant challenges in development, including availability of clinical data to algorithmically derive clinical decision support, as well as trustworthiness of the evaluated treatment options. We propose a collaborative approach that mitigates the regulatory and ethical concerns that are broadly discussed when automated decision-making tools are to be included into clinical routine.

Making complex medical decisions is becoming an increasingly challenging task due to the growing amount of available evidence to consider and the higher demand for personalized treatment and patient care. IT systems for the provision of clinical decision support (CDS) can provide sustainable relief if decisions are automatically evaluated and processed. In this paper, we propose an approach for quantifying similarity between new and previously recorded medical cases to enable significant knowledge transfer for reasoning tasks on a patient-level. Methodologically, 102 medical cases with oropharyngeal carcinoma were analyzed retrospectively. Based on independent disease characteristics, patient-specific data vectors including relevant information entities for primary and adjuvant treatment decisions were created. Utilizing the ϕK correlation coefficient as the methodological foundation of our approach, we were able to determine the predictive impact of each characteristic, thus enabling significant reduction of the feature space to allow for further analysis of the intra-variable distances between the respective feature states. The results revealed a significant feature-space reduction from initially 19 down to only 6 diagnostic variables (ϕK correlation coefficient ≥ 0.3, ϕK significance test ≥ 2.5) for the primary and 7 variables (from initially 14) for the adjuvant treatment setting. Further investigation on the resulting characteristics showed a non-linear behavior in relation to the corresponding distances on intra-variable level. Through the implementation of a 10-fold cross-validation procedure, we were further able to identify 8 (primary treatment) matching cases with an evaluation score of 1.0 and 9 (adjuvant treatment) matching cases with an evaluation score of 0.957 based on their shared treatment procedure as the endpoint for similarity definition. Based on those promising results, we conclude that our proposed method for using data-driven similarity measures for application in medical decision-making is able to offer valuable assistance for physicians. Furthermore, we consider our approach as universal in regard to other clinical use-cases, which would allow for an easy-to-implement adaptation for a range of further medical decision-making scenarios.

Chimeric Antigen Receptor (CAR) T cell therapy revolutionized treatment for relapsed/refractory multiple myeloma (RRMM). The standard lymphodepletion (LDP) regimen is fludarabine and cyclophosphamide (Flu/Cy), but bendamustine has shown comparable efficacy with lower toxicity and is a potentially safer option for patients with renal impairment. This study retrospectively analyzes clinical outcomes and CAR T-cell dynamics in patients with RRMM receiving idecabtagene vicleucel (Ide-cel) or ciltacabtagene autoleucel (Cilta-cel), with a particular focus on the impact of renal function and LDP regimen. We included a total of 87 patients who received CAR T-cell therapy: 53 with none/mild chronic kidney disease (CKD) and 34 with moderate/severe CKD. LDP consisted of standard-dose Flu/Cy in 67 patients, reduced-dose Flu/Cy in 11 patients, and bendamustine in 9 patients. Toxicity outcomes were monitored over more than 2,000 patient-days, capturing over 20,000 individual data points. Patients with moderate/severe CKD received standard Flu/Cy conditioning less frequently (55.9% vs. 96.2%, p<0.001) and bendamustine-based LDP more often (26.5% vs. 0%, p<0.001). In the survival analysis, we found that CKD status alone did not significantly impact outcomes (OS: p=0.93; PFS: p=0.82). Similarly, when stratified by CAR T-cell product, CKD status did not impact PFS in Ide-cel recipients (p=0.46). In contrast, among Cilta-cel-treated patients, those with none/mild CKD demonstrated significantly improved PFS (p=0.026). LDP regimen had no effect on PFS (p=0.21). In Firth's penalized Cox model, neither CKD nor LDP independently predicted PFS; however, a significant interaction between CKD status and CAR T product was observed (HR 19.32, 95% CI 1.25-2176, p=0.03). Subgroup analyses confirmed that Cilta-cel conferred superior PFS compared with Ide-cel in patients with none/mild CKD (p<0.001), whereas no benefit was seen in those with moderate/severe CKD. The median absolute lymphocyte count (ALC) nadir was observed on day -1 in patients receiving Flu/Cy (0.033 × 10⁹/L) and on day +1 in those receiving bendamustine (0.059 × 10⁹/L). At the time of CAR T-cell infusion, median ALC was significantly higher in the bendamustine group compared to the complete dosage Flu/Cy cohort (p = 0.03). Overall CAR T-cell expansion kinetics did not differ according to CKD status or LDP regimen. Notably, at day 7, Cilta-Cel patients had significantly higher CD4 CAR T cell percentages (of total CD3 CAR T cells) compared to Ide-Cel in the none/mild CKD group (p < 0.001), a difference that persisted through days 14 (p<0.001) and 30 (p<0.001). Analysis of hematologic toxicities showed that bendamustine-based LDP was associated with lower rates of early N-ICAHT (p=0.002) and a higher ANC nadir (p<0.001), while rates of CRS and ICANS were similar between groups. These findings provide a signal supporting the feasibility and safety of bendamustine-based lymphodepletion and highlight the prognostic relevance of renal function in Cilta-Cel-treated patients. While Cilta-Cel generally confers superior efficacy compared with Ide-Cel, this advantage may be attenuated in patients with moderate/severe CKD, potentially due to competing risks such as frailty, early toxicity, or impaired cellular fitness.