Novel Mouse Avatars Could Advance Personalized Medicine For Cancer

By Deborah Borfitz 

May 3, 2023 | Up to now, mouse avatars haven’t proven to be particularly good tools for forecasting treatment response in human cancer patients. But new models being generated by an international team of researchers are expected to be far more predictive of what is going to happen with different cohorts of clinical trial participants, according to José Ángel Martínez-Climent, M.D., Ph.D., senior scientist at Cima University of Navarra (Spain) and professor of medicine in the university’s school of medicine. 

Their most recently studied mouse avatars, designed to mimic the development and evolution of multiple myeloma, were described in an article that was recently published in Nature Medicine (DOI: 10.1038/s41591-022-02178-3). The artificial mice were shown to accurately reflect key aspects of the disease in patients, in terms of its progression and of predicting response to immunotherapy drug combinations. 

The research team for the multicenter study included representatives from Switzerland (Roche Pharmaceutical Innovation Center), the United States (Mayo Clinic, Yale University, Dana-Farber Cancer Institute, Cornell University, and Genentech), and Japan (University of Tsukuba), as well as several research centers in Spain. Mouse models recapitulated the inception of multiple myeloma not only from the genetic but also the cellular, tumor, and immunological points of view, says Martínez-Climent. 

Notably, the mouse avatars captured disease development from the very beginning and, it is hoped, will provide needed tools for anticipating how multiple myeloma patients will fare on investigational therapies, he says. Ultimately, it is expected that the avatars will enable personalized treatments for the blood cancer that occurs in the bone marrow and is the second most frequent hematological cancer and typically incurable. 

Creation of the mouse avatars required genetic engineering and multi-omic cellular and molecular technologies to analyze over 500 genetically heterogenous mice that developed myeloma as well as tumor cells from more than 1,000 patients with the disease. The samples came from individuals being seen in the hematology department at the University of Navarra. 

For this study, researchers generated 15 mouse models that reflect the primary clinical, genetic, and immunological characteristics of multiple myeloma seen in patients, Martínez-Climent reports. Work on avatars for other hematological malignancies such as lymphomas and more prevalent solid tumors of the lung, colon, and breast is now underway. 

Big pharma has taken note, as evidenced by the dozen research agreements with study sponsors that have already been signed, he continues. Two preclinical trials are now being conducted using the novel mouse models—one for multiple myeloma and the other for lymphoma—in parallel with clinical trials in patients and following the same protocol in terms of drug dosing (adjusted to the mouse weight), scheduling, and study procedures (bloodwork and micro computed tomography) for measuring response to treatment using non-invasive methods, he says. 

The objective is to apply what is learned from the mouse avatars to optimize the design of human studies, improving their success rate and lowering the cost of bringing life-enhancing medicines to market. In the latest preclinical study, for example, the suggestion is that an elevated ratio of tumor-reactive CD8+ T cells to immunosuppressive Treg cells in the bone marrow microenvironment is a predictor of immunotherapy responses, particularly to PD-1/PD-L1 inhibitors, in multiple myeloma.  

However, as Martínez-Climent points out, only 15% of patients with newly diagnosed myeloma show an elevated CD8/Treg cell ratio that predicts response to PD-1/PD-L1 inhibitors. These results may provide a scientific explanation to why phase III clinical trials using PD-1 inhibitors have failed, he says. 

‘Absolutely Unique’ 

The mouse avatars deployed in the preclinical study collectively account for the most common somatic mutations known to be involved in initiating multiple myeloma, says Martínez-Climent. Unlike traditional avatars that are based on tumor cells transplanted from patients into immune-deficient mice, the new genetically modified models reproduce the disease using only mouse cells. 

This makes them “absolutely unique,” he says. With the previous models, the transferred cells could not be transformed because they were already malignant. 

With the new avatars, the disease originates in mice and the process can take six months to a year—up to one-third their average lifespan, or the equivalent of 30 to 40 years of tumor development in humans, Martínez-Climent says. Cancers can be characterized from “initiation of the disease to the end” when it has progressed to the point of killing the mice. 

Although it is feasible for this sort of work to be done elsewhere, neither manipulating the genome of the mice to create the avatars nor using them to conduct preclinical trials is an easy feat, he adds. “These are very complex and laborious experiments to perform.”  

The avatars are allowing researchers to mimic in the laboratory clinical situations associated with the worst outcomes (including high-risk multiple myeloma, acquired therapeutic resistance, and extramedullary disease) to advance the investigation of new therapeutic strategies, he notes. Additionally, novel therapies can be tested both in the early precursor stages of the disease (i.e., smoldering multiple myeloma and monoclonal gammopathy of undetermined significance) when cancer cells might be most vulnerable and in the minimal residual disease state after treatment when few tumor cells remain. 

If researchers succeed in validating the information provided by their preclinical models with patient data, says Martínez-Climent, they could prove useful on their own as sponsors are designing their clinical trials—ultimately, for many different types of tumors. Multiple drug combinations might be tested in tumor-specific avatars, for instance, to pick the ones demonstrating the best treatment responses.