Designing an effective, tumor-specific, and safe anti-cancer immunotherapy is similar to solving a complex puzzle. As many pieces are scattered, putting them together often demands a collaborative effort. We undertook such an effort as a Polish-Norwegian consortium, primarily involving the Medical University of Warsaw and Oslo University Hospital, carried out under the ALTERCAR project.

For decades, scientists have been trying to empower the human body to fight cancer, relying on the host’s immune defenses. One of the greatest achievements was the modification of cytotoxic T cells with a chimeric antigen receptor (CAR), enabling specific tumor cell recognition and killing. The first approved CAR-T cell therapy was directed against the CD19 surface protein, which is present on all normal B-cells and B-cells that underwent malignant transformation, giving rise to B-cell acute lymphoblastic leukemia (B-ALL) and B-cell lymphoma. Even though patients with such conditions are well managed with standard chemo-immunotherapy, especially children, those who develop refractory or relapsed disease have much poorer outcomes. Luckily, thanks to cellular immunotherapy, many refractory or relapsed B-ALL/lymphoma patients, who had low chances of finding successful treatment before the implementation of CD19 CAR-T, survive. Why, then, focus on finding novel targets and invest in further CAR-T development?

Like all anti-cancer treatments, therapy with CAR-T cells is not 100% effective. The challenges ahead of successful treatment are multifaceted, and malignant cells constantly seek a way to escape even the most elegant therapy, such as CD19 CAR-T. In particular, it turned out that malignant B-cells can escape the CAR-T cell-induced death through the loss or decrease of the CD19 target antigen. Such observation prompted further research to find novel targets that could be used in patients who did not respond or relapsed after CD19 CAR-T treatment.

Our journey to developing an alternative CAR started with the optimization of the proper tools and models. While standard methods for target identification rely on analysis of cancer cell lines, we decided to perform proteomic analysis of patient-derived cells to preserve cellular heterogeneity. However, as good as it sounds, surface protein analysis of biotinylated proteins by quantitative mass spectrometry requires a high number of cells, therefore, we propagated human B-ALL cells in immunodeficient mice, creating patient-derived xenografts.

As it turned out, this was just the start of putting the puzzle pieces together.

Analyzing high-throughput data can be overwhelming, to say the least. But what exactly were we looking for in our dataset? There are many ways to define an ideal CAR target. For one, it should exhibit high and uniform expression on cancer cells while being absent on vital tissues or cells responsible for normal hematopoiesis. Importantly, given that ideal is a subjective term, not all (if any) identified targets would be exclusively expressed by cancer cells. With this in mind, we narrowed down our initial pool of more than 1000 membrane proteins and identified 18 potential CAR targets, finally choosing LILRB1 as the one that met most of our stringent criteria.

Though we had completed the crucial step of selecting the CAR target, one could argue that this was the easier part. Indeed, the greatest challenge in developing a new CAR is proving its safety. We therefore assessed both the mRNA and protein levels of LILRB1 in normal cells and found it to be predominantly expressed in B-cells and monocytes, suggesting the feasibility of LILRB1 CAR-T cell manufacturing and a low likelihood of overt toxicity. While motivated by these observations, we were still eager to address the remaining key questions: 1) Is LILRB1 expressed by different types of malignant cells, or only by the cells we used in our screening? and 2) Is LILRB1 maintained in patients who have previously undergone CD19-targeted immunotherapy? Remarkably, we detected LILRB1 expression in a cohort of B-ALL and B-cell lymphoma patients, and most strikingly, we found that its expression is sustained in patients who had previously undergone CD19-directed treatment.

With a promising candidate for ALTERCAR therapy, we moved on to CAR design. Based on two clones of murine anti-LILRB1 antibodies, we developed two distinct LILRB1 CARs, but only one showed robust efficacy against B-ALL and B-cell lymphoma. Even though our search for new CAR targets initially concentrated on B-cell-derived neoplasms, we were intrigued by the high expression of LILRB1 on normal monocytes and hypothesized that it might also be found on monocyte-derived cancers. Indeed, we found robust expression of LILRB1 in primary monocytic acute myeloid leukemia (AML), paving the way for potential LILRB1 CAR-T therapy in this cancer.

Why does our discovery matter? It opens up a range of therapeutic possibilities for blood cancers, from resistant B-cell cancers to myeloid leukemias. LILRB1 CAR-T cells offer a promising alternative for patients who relapse after previous immunotherapies, targeting CD20, CD22, or CD19, including those with CD19-negative relapses following CD19 CAR-T therapy. Additionally, LILRB1 CAR-T cells could address the challenge of lymphoid-to-myeloid lineage switching in B-ALL, providing a treatment option when B-cell markers are lost during treatment. Furthermore, LILRB1 CAR-T cells have the potential to extend beyond B-cell malignancies, offering hope for patients with challenging monocytic leukemias, where safe and effective CAR targets remain undiscovered, and currently tested CAR-T therapies pose a risk of severe side effects like bone marrow toxicity. While much research remains, we are excited to continue the ALTERCAR project and advance studies on LILRB1 CAR-T cells, with the ultimate goal of translating our findings into clinical practice.

 ________________________________________

The project was funded by the National Centre for Research and Development, Poland.