Myeloma Cell Dynamics in Response to Treatment Supports a Model of Hierarchical Differentiation and Clonal Evolution

Longitudinal efficacy and safety modeling and simulation framework to aid dose selection of belantamab mafodotin for patients with multiple myeloma

Belantamab mafodotin, a monomethyl auristatin F (MMAF)-containing monoclonal antibody-drug conjugate (ADC), demonstrated deep and durable responses in the DRiving Excellence in Approaches to Multiple Myeloma (DREAMM)-1 and pivotal DREAMM-2 studies in patients with relapsed/refractory multiple myeloma. As with other MMAF-containing ADCs, ocular adverse events were observed. To predict the effects of belantamab mafodotin dosing regimens and dose-modification strategies on efficacy and ocular safety end points, DREAMM-1 and DREAMM-2 data across a range of doses were used to develop an integrated simulation framework incorporating two separate longitudinal models and the published population pharmacokinetic model. A concentration-driven tumor growth inhibition model described the time course of serum M-protein concentration, a measure of treatment response, whereas a discrete time Markov model described the time course of ocular events graded with the GSK Keratopathy and Visual Acuity scale. Significant covariates included baseline β2 -microglobulin on growth rate, baseline M-protein on kill rate, extramedullary disease on the effect compartment rate constant, and baseline soluble B cell maturation antigen on maximal effect. Efficacy and safety end points were simulated for various doses with dosing intervals of 1, 3, 6, and 9 weeks and various event-driven dose-modification strategies. Simulations predicted that lower doses and longer dosing intervals were associated with lower probability and lower overall time with Grade 3+ and Grade 2+ ocular events compared with the reference regimen (2.5 mg/kg every 3 weeks), with a less-than-proportional reduction in efficacy. The predicted improved benefit-risk profiles of certain dosing schedules and dose modifications from this integrated framework has informed trial designs for belantamab mafodotin, supporting dose-optimization strategies.

Machine Learning Approach for Rapid, Accurate Point-of-Care Prediction of M-Spike Values in Multiple Myeloma

PURPOSE

The gold standard for monitoring response status in patients with multiple myeloma (MM) is serum and urine protein electrophoresis which quantify M-spike proteins; however, the turnaround time for results is 3-7 days which delays treatment decisions. We hypothesized that machine learning (ML) could integrate readily available clinical and laboratory data to rapidly and accurately predict patient M-spike values.

METHODS

A retrospective chart review was performed using the deidentified, electronic medical records of 171 patients with MM.

RESULTS

Random forest (RF) analysis identified the weighted value of each independent variable (N = 43) integrated into the ML algorithm. Pearson and Spearman coefficients indicated that the ML-predicted M-spike values correlated highly with laboratory-measured serum protein electrophoresis values. Feature selected RF modeling revealed that only two variables-the first lagged M-spike and serum total protein-accurately predicted the M-spike.

CONCLUSION

Taken together, our results demonstrate the feasibility and prognostic potential of ML tools that integrate electronic data to longitudinally monitor disease burden. ML tools support the seamless, secure exchange of patient information to expedite and personalize clinical decision making and overcome geographic, financial, and social barriers that currently limit the access of underserved populations to cancer care specialists so that the benefits of medical progress are not limited to selected groups.

Phenotypic deconvolution in heterogeneous cancer cell populations using drug-screening data

Tumor heterogeneity is an important driver of treatment failure in cancer since therapies often select for drug-tolerant or drug-resistant cellular subpopulations that drive tumor growth and recurrence. Profiling the drug-response heterogeneity of tumor samples using traditional genomic deconvolution methods has yielded limited results, due in part to the imperfect mapping between genomic variation and functional characteristics. Here, we leverage mechanistic population modeling to develop a statistical framework for profiling phenotypic heterogeneity from standard drug-screen data on bulk tumor samples. This method, called PhenoPop, reliably identifies tumor subpopulations exhibiting differential drug responses and estimates their drug sensitivities and frequencies within the bulk population. We apply PhenoPop to synthetically generated cell populations, mixed cell-line experiments, and multiple myeloma patient samples and demonstrate how it can provide individualized predictions of tumor growth under candidate therapies. This methodology can also be applied to deconvolution problems in a variety of biological settings beyond cancer drug response.

Validation of a Mathematical Model Describing the Dynamics of Chemotherapy for Chronic Lymphocytic Leukemia In Vivo

In recent years, mathematical models have developed into an important tool for cancer research, combining quantitative analysis and natural processes. We have focused on Chronic Lymphocytic Leukemia (CLL), since it is one of the most common adult leukemias, which remains incurable. As the first step toward the mathematical prediction of in vivo drug efficacy, we first found that logistic growth best described the proliferation of fluorescently labeled murine A20 leukemic cells injected in immunocompetent Balb/c mice. Then, we tested the cytotoxic efficacy of Ibrutinib (Ibr) and Cytarabine (Cyt) in A20-bearing mice. The results afforded calculation of the killing rate of the A20 cells as a function of therapy. The experimental data were compared with the simulation model to validate the latter’s applicability. On the basis of these results, we developed a new ordinary differential equations (ODEs) model and provided its sensitivity and stability analysis. There was excellent accordance between numerical simulations of the model and results from in vivo experiments. We found that simulations of our model could predict that the combination of Cyt and Ibr would lead to approximately 95% killing of A20 cells. In its current format, the model can be used as a tool for mathematical prediction of in vivo drug efficacy, and could form the basis of software for prediction of personalized chemotherapy.

The role of E3 ubiquitin ligase in multiple myeloma: potential for cereblon E3 ligase modulators in the treatment of relapsed/refractory disease

INTRODUCTION

Insights into the mechanisms of protein homeostasis and proteasomal degradation have led to new strategies of redirecting the ubiquitin-proteasome system (UPS) to reduce or eliminate proteins or survival factors key to malignant pathobiology, multiple myeloma (MM) in particular. These strategies have enabled researchers to target proteins that were previously considered difficult to modulate by pharmacological means.

AREAS COVERED

This review provides a brief overview of UPS biology, particularly the role of the CRL4^CRBN E3 ubiquitin ligase complex, and summarizes current strategies for co-opting the UPS, including CELMoD compounds, SNIPERs, PROTACs, and degronimids. A detailed discussion is provided on lead CELMoD compounds iberdomide and mezigdomide, which are currently being evaluated in clinical trials in patients with MM.

EXPERT OPINION

Since a high proportion of patients develop drug resistance, it is vital to have novel therapeutic agents for treating relapsed patients with MM more effectively. It is encouraging that the expanding pathophysiological insight into cellular signaling pathways in MM increasingly translates into the development of novel therapeutic agents such as targeted protein degraders. This holds promise for improving outcomes in MM and beyond.

False positive results: a challenge for laboratory physicians and hematologists in treating multiple myeloma with daratumumab

Daratumumab injection was approved by China in 2019 for the treatment of recurrent or refractory multiple myeloma. However, the molecular weight of daratumumab, an immunoglobin G1 kappa human monoclonal antibody, was similar to that of M protein and could not be distinguished from IgG κ M protein in SPEP and SIFE. It might lead to false-positive detection resulting in misdiagnose and confusing evaluation of therapeutic response, especially for patients with IgG κ M proteins. Herein, we reported two cases encountered in our daily clinical work. These two case reports could serve as a reminder to global hematologists who have not yet started or just begun to use the drug of daratumumab.

Experimental Validation of a Mathematical Model to Describe the Drug Cytotoxicity of Leukemic Cells

Chlorambucil (Chl), Melphalan (Mel), and Cytarabine (Cyt) are recognized drugs used in the chemotherapy of patients with advanced Chronic Lymphocytic Leukemia (CLL). The optimal treatment schedule and timing of Chl, Mel, and Cyt administration remains unknown and has traditionally been decided empirically and independently of preclinical in vitro efficacy studies. As a first step toward mathematical prediction of in vivo drug efficacy from in vitro cytotoxicity studies, we used murine A20 leukemic cells as a test case of CLL. We first found that logistic growth best described the proliferation of the cells in vitro. Then, we tested in vitro the cytotoxic efficacy of Chl, Mel, and Cyt against A20 cells. On the basis of these experimental data, we found the parameters for cancer cell death rates that were dependent on the concentration of the respective drugs and developed a mathematical model involving nonlinear ordinary differential equations. For the proposed mathematical model, three equilibrium states were analyzed using the general method of Lyapunov, with only one equilibrium being stable. We obtained a very good symmetry between the experimental results and numerical simulations of the model. Our novel model can be used as a general tool to study the cytotoxic activity of various drugs with different doses and modes of action by appropriate adjustment of the values for the selected parameters.

Employment of Artificial Intelligence Based on Routine Laboratory Results for the Early Diagnosis of Multiple Myeloma

Objective

In order to enhance the detection rate of multiple myeloma and execute an early and more precise disease management, an artificial intelligence assistant diagnosis system is developed.

Methods

4,187 routine blood and biochemical examination records were collected from Shengjing Hospital affiliated to China Medical University from January 2010 to January 2020, which include 1,741 records of multiple myeloma (MM) and 2,446 records of non-myeloma (infectious diseases, rheumatic immune system diseases, hepatic diseases and renal diseases). The data set was split into training and test subsets with the ratio of 4:1 while connecting hemoglobin, serum creatinine, serum calcium, immunoglobulin (A, G and M), albumin, total protein, and the ratio of albumin to globulin data. An early assistant diagnostic model of MM was established by Gradient Boosting Decision Tree (GBDT), Support Vector Machine (SVM), Deep Neural Networks (DNN), and Random Forest (RF). Out team calculated the precision and recall of the system. The performance of the diagnostic model was evaluated by using the receiver operating characteristic (ROC) curve.

Results

By designing the features properly, the typical machine learning algorithms SVM, DNN, RF and GBDT all performed well. GBDT had the highest precision (92.9%), recall (90.0%) and F1 score (0.915) for the myeloma group. The maximized area under the ROC (AUROC) was calculated, and the results of GBDT (AUC: 0.975; 95% confidence interval (CI): 0.963–0.986) outperformed that of SVM, DNN and RF.

Conclusion

The model established by artificial intelligence derived from routine laboratory results can accurately diagnose MM, which can boost the rate of early diagnosis.

Reconstructing tumor evolutionary histories and clone trees in polynomial-time with SubMARine

Tumors contain multiple subpopulations of genetically distinct cancer cells. Reconstructing their evolutionary history can improve our understanding of how cancers develop and respond to treatment. Subclonal reconstruction methods cluster mutations into groups that co-occur within the same subpopulations, estimate the frequency of cells belonging to each subpopulation, and infer the ancestral relationships among the subpopulations by constructing a clone tree. However, often multiple clone trees are consistent with the data and current methods do not efficiently capture this uncertainty; nor can these methods scale to clone trees with a large number of subclonal populations.

Here, we formalize the notion of a partially-defined clone tree (partial clone tree for short) that defines a subset of the pairwise ancestral relationships in a clone tree, thereby implicitly representing the set of all clone trees that have these defined pairwise relationships. Also, we introduce a special partial clone tree, the Maximally-Constrained Ancestral Reconstruction (MAR), which summarizes all clone trees fitting the input data equally well. Finally, we extend commonly used clone tree validity conditions to apply to partial clone trees and describe SubMARine, a polynomial-time algorithm producing the subMAR, which approximates the MAR and guarantees that its defined relationships are a subset of those present in the MAR. We also extend SubMARine to work with subclonal copy number aberrations and define equivalence constraints for this purpose. Further, we extend SubMARine to permit noise in the estimates of the subclonal frequencies while retaining its validity conditions and guarantees. In contrast to other clone tree reconstruction methods, SubMARine runs in time and space that scale polynomially in the number of subclones.

We show through extensive noise-free simulation, a large lung cancer dataset and a prostate cancer dataset that the subMAR equals the MAR in all cases where only a single clone tree exists and that it is a perfect match to the MAR in most of the other cases. Notably, SubMARine runs in less than 70 seconds on a single thread with less than one Gb of memory on all datasets presented in this paper, including ones with 50 nodes in a clone tree. On the real-world data, SubMARine almost perfectly recovers the previously reported trees and identifies minor errors made in the expert-driven reconstructions of those trees.

The freely-available open-source code implementing SubMARine can be downloaded at https://github.com/morrislab/submarine.

Cancer cells accumulate mutations over time and consist of genetically distinct subpopulations. Their evolutionary history (as represented by tumor phylogenies) can be inferred from bulk cancer genome sequencing data. Current tumor phylogeny reconstruction methods have two main issues: they are slow, and they do not efficiently represent uncertainty in the reconstruction.

To address these issues, we developed SubMARine, a fast algorithm that summarizes all valid phylogenies in an intuitive format. SubMARine solved all reconstruction problems in this manuscript in less than 70 seconds, orders of magnitude faster than other methods. These reconstruction problems included those with up to 50 subclones; problems that are too large for other algorithms to even attempt. SubMARine achieves these result because, unlike other algorithms, it performs its reconstruction by identifying an upper-bound on the solution set of trees and the amount of noise in the estimates of the subclonal frequencies. In the vast majority of cases we checked, i. e. an extensive noise-free simulation, a lung cancer and a prostate cancer dataset, this upper bound is tight: when only a single solution exists, SubMARine converges to it every time. When multiple solutions exist, our algorithm correctly recovers the uncertain relationships in 71% of cases.

In addition to solving these two major challenges, we introduce some useful new concepts for and open research problems in the field of tumor phylogeny reconstruction. Specifically, we formalize the concept of a partial clone tree which provides a set of constraints on the solution set of clone trees; and provide a complete set of conditions under which a partial clone tree is valid. These conditions guarantee that all trees in the solution set satisfy the constraints implied by the partial clone tree.

Early M-Protein Dynamics Predicts Progression-Free Survival in Patients With Relapsed/Refractory Multiple Myeloma

This study aimed to predict long‐term progression‐free survival (PFS) using early M‐protein dynamic measurements in patients with relapsed/refractory multiple myeloma (MM). The PFS was modeled based on dynamic M‐protein data from two phase III studies, POLLUX and CASTOR, which included 569 and 498 patients with relapsed/refractory MM, respectively. Both studies compared active controls (lenalidomide and dexamethasone, and bortezomib and dexamethasone, respectively) alone vs. in combination with daratumumab. Three M‐protein dynamic features from the longitudinal M‐protein data were evaluated up to different time cutoffs (1, 2, 3, and 6 months). The abilities of early M‐protein dynamic measurements to predict the PFS were evaluated using Cox proportional hazards survival models. Both univariate and multivariable analyses suggest that maximum reduction of M‐protein (i.e., depth of response) was the most predictive of PFS. Despite the statistical significance, the baseline covariates provided very limited predictive value regarding the treatment effect of daratumumab. However, M‐protein dynamic features obtained within the first 2 months reasonably predicted PFS and the associated treatment effect of daratumumab. Specifically, the areas under the time‐varying receiver operating characteristic curves for the model with the first 2 months of M‐protein dynamic data were ~ 0.8 and 0.85 for POLLUX and CASTOR, respectively. Early M‐protein data within the first 2 months can provide a prospective and reasonable prediction of future long‐term clinical benefit for patients with MM.

Predicting circulating biomarker response and its impact on the survival of advanced melanoma patients treated with adjuvant therapy

Advanced melanoma remains a disease with poor prognosis. Several serologic markers have been investigated to help monitoring and prognostication, but to date only lactate dehydrogenase (LDH) has been validated as a standard prognostic factor biomarker for this disease by the American Joint Committee on Cancer. In this work, we built a semi-mechanistic model to explore the relationship between the time course of several circulating biomarkers and overall or progression free survival in advanced melanoma patients treated with adjuvant high-dose interferon-\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\boldsymbol{\alpha }}{\bf{2}}{\bf{b}}$$\end{document}α2b. Additionally, due to the adverse interferon tolerability, a semi-mechanistic model describing the side effects of the treatment in the absolute neutrophil counts is proposed in order to simultaneously analyze the benefits and toxic effects of this treatment. The results of our analysis suggest that the relative change from baseline of LDH was the most significant predictor of the overall survival of the patients. Unfortunately, there was no significant difference in the proportion of patients with elevated serum biomarkers between the patients who recurred and those who remained free of disease. Still, we believe that the modelling framework presented in this work of circulating biomarkers and adverse effects could constitute an additional strategy for disease monitoring in advance melanoma patients.

Upregulation of FOXM1 leads to diminished drug sensitivity in myeloma

BACKGROUND

Following up on previous work demonstrating the involvement of the transcription factor forkhead box M1 (FOXM1) in the biology and outcome of a high-risk subset of newly diagnosed multiple myeloma (nMM), this study evaluated whether FOXM1 gene expression may be further upregulated upon tumor recurrence in patients with relapsed multiple myeloma (rMM). Also assessed was the hypothesis that increased levels of FOXM1 diminish the sensitivity of myeloma cells to commonly used myeloma drugs, such as the proteasome inhibitor bortezomib (Bz) and the DNA intercalator doxorubicin (Dox).

METHODS

FOXM1 message was evaluated in 88 paired myeloma samples from patients with nMM and rMM, using gene expression microarrays as measurement tool. Sources of differential gene expression were identified and outlier analyses were performed using statistical methods. Two independent human myeloma cell lines (HMCLs) containing normal levels of FOXM1 (FOXM1N) or elevated levels of lentivirus-encoded FOXM1 (FOXM1Hi) were employed to determine FOXM1-dependent changes in cell proliferation, survival, efflux-pump activity, and drug sensitivity. Levels of retinoblastoma (Rb) protein were determined with the assistance of Western blotting.

RESULTS

Upregulation of FOXM1 occurred in 61 of 88 (69%) patients with rMM, including 4 patients that exhibited > 20-fold elevated expression peaks. Increased FOXM1 levels in FOXM1Hi myeloma cells caused partial resistance to Bz (1.9-5.6 fold) and Dox (1.5-2.9 fold) in vitro, using FOXM1N myeloma as control. Reduced sensitivity of FOXM1Hi cells to Bz was confirmed in vivo using myeloma-in-mouse xenografts. FOXM1-dependent regulation of total and phosphorylated Rb agreed with a working model of myeloma suggesting that FOXM1 governs both chromosomal instability (CIN) and E2F-dependent proliferation, using a mechanism that involves interaction with NIMA related kinase 2 (NEK2) and cyclin dependent kinase 6 (CDK6), respectively.

CONCLUSIONS

These findings enhanced our understanding of the emerging FOXM1 genetic network in myeloma and provided preclinical support for the therapeutic targeting of the FOXM1-NEK2 and CDK4/6-Rb-E2F pathways using small-drug CDK and NEK2 inhibitors. Clinical research is warranted to assess whether this approach may overcome drug resistance in FOXM1Hi myeloma and, thereby, improve the outcome of patients in which the transcription factor is expressed at high levels.

Prevention Is the Best Treatment: The Case for Understanding the Transition from Monoclonal Gammopathy of Undetermined Significance to Myeloma

Multiple myeloma is an invariably fatal cancer of plasma cells. Despite tremendous advances in treatment, this malignancy remains incurable in most individuals. We postulate that strategies aimed at prevention have the potential to be more effective in preventing myeloma-related death than additional pharmaceutical strategies aimed at treating advanced disease. Here, we present a rationale for the development of prevention therapy and highlight potential target areas of study.

A new conceptual framework for the therapy by optimized multidimensional pulses of therapeutic activity. The case of multiple myeloma model

We developed simulation methodology to assess eventual therapeutic efficiency of exogenous multiparametric changes in a four-component cellular system described by the system of ordinary differential equations. The method is numerically implemented to simulate the temporal behavior of a cellular system of multiple myeloma cells. The problem is conceived as an inverse optimization task where the alternative temporal changes of selected parameters of the ordinary differential equations represent candidate solutions and the objective function quantifies the goals of the therapy. The system under study consists of two main cellular components, tumor cells and their cellular environment, respectively. The subset of model parameters closely related to the environment is substituted by exogenous time dependencies - therapeutic pulses combining continuous functions and discrete parameters subordinated thereafter to the optimization. Synergistic interaction of temporal parametric changes has been observed and quantified whereby two or more dynamic parameters show effects that absent if either parameter is stimulated alone. We expect that the theoretical insight into unstable tumor growth provided by the sensitivity and optimization studies could, eventually, help in designing combination therapies.

Methods for determining key components in a mathematical model for tumor-immune dynamics in multiple myeloma

In this work, we analyze a mathematical model we introduced previously for the dynamics of multiple myeloma and the immune system. We focus on four main aspects: (1) obtaining and justifying ranges and values for all parameters in the model; (2) determining a subset of parameters to which the model is most sensitive; (3) determining which parameters in this subset can be uniquely estimated given certain types of data; and (4) exploring the model numerically. Using global sensitivity analysis techniques, we found that the model is most sensitive to certain growth, loss, and efficacy parameters. This analysis provides the foundation for a future application of the model: prediction of optimal combination regimens in patients with multiple myeloma.

Extinction rates in tumour public goods games

Cancer evolution and progression are shaped by cellular interactions and Darwinian selection. Evolutionary game theory incorporates both of these principles, and has been proposed as a framework to understand tumour cell population dynamics. A cornerstone of evolutionary dynamics is the replicator equation, which describes changes in the relative abundance of different cell types, and is able to predict evolutionary equilibria. Typically, the replicator equation focuses on differences in relative fitness. We here show that this framework might not be sufficient under all circumstances, as it neglects important aspects of population growth. Standard replicator dynamics might miss critical differences in the time it takes to reach an equilibrium, as this time also depends on cellular turnover in growing but bounded populations. As the system reaches a stable manifold, the time to reach equilibrium depends on cellular death and birth rates. These rates shape the time scales, in particular, in coevolutionary dynamics of growth factor producers and free-riders. Replicator dynamics might be an appropriate framework only when birth and death rates are of similar magnitude. Otherwise, population growth effects cannot be neglected when predicting the time to reach an equilibrium, and cell-type-specific rates have to be accounted for explicitly.

Identification of precision treatment strategies for relapsed/refractory multiple myeloma by functional drug sensitivity testing

Novel agents have increased survival of multiple myeloma (MM) patients, however high-risk and relapsed/refractory patients remain challenging to treat and their outcome is poor. To identify novel therapies and aid treatment selection for MM, we assessed the ex vivo sensitivity of 50 MM patient samples to 308 approved and investigational drugs. With the results we i) classified patients based on their ex vivo drug response profile; ii) identified and matched potential drug candidates to recurrent cytogenetic alterations; and iii) correlated ex vivo drug sensitivity to patient outcome. Based on their drug sensitivity profiles, MM patients were stratified into four distinct subgroups with varied survival outcomes. Patients with progressive disease and poor survival clustered in a drug response group exhibiting high sensitivity to signal transduction inhibitors. Del(17p) positive samples were resistant to most drugs tested with the exception of histone deacetylase and BCL2 inhibitors. Samples positive for t(4;14) were highly sensitive to immunomodulatory drugs, proteasome inhibitors and several targeted drugs. Three patients treated based on the ex vivo results showed good response to the selected treatments. Our results demonstrate that ex vivo drug testing may potentially be applied to optimize treatment selection and achieve therapeutic benefit for relapsed/refractory MM.

Hierarchical tissue organization as a general mechanism to limit the accumulation of somatic mutations

How can tissues generate large numbers of cells, yet keep the divisional load (the number of divisions along cell lineages) low in order to curtail the accumulation of somatic mutations and reduce the risk of cancer? To answer the question we consider a general model of hierarchically organized self-renewing tissues and show that the lifetime divisional load of such a tissue is independent of the details of the cell differentiation processes, and depends only on two structural and two dynamical parameters. Our results demonstrate that a strict analytical relationship exists between two seemingly disparate characteristics of self-renewing tissues: divisional load and tissue organization. Most remarkably, we find that a sufficient number of progressively slower dividing cell types can be almost as efficient in minimizing the divisional load, as non-renewing tissues. We argue that one of the main functions of tissue-specific stem cells and differentiation hierarchies is the prevention of cancer.

To limit the accumulation of somatic mutations, renewing tissues must minimize the number of times each cell divides during differentiation. Here, the authors analytically derive the lower limit of lifetime divisional load of a tissue, show that hierarchically differentiating tissues can approach this limit, and that this depends on uneven divisional rates across the hierarchy.