Increasing the number of target antigens beyond bispecific CAR-T cells should further reduce the risk of antigen escape and thus of disease relapse. In vitro and in vivo experiments demonstrated increased CAR-T cell expansion and improved antitumor effects through the sequential administration of single targeting anti-CD19, anti-CD20 and anti-CD22 cells. Compassionate treatment of two patients with Burkitt lymphoma and LBCL seems to confirm the observation, showing delayed re-expansion of previously infused cells upon the administration of subsequent products. CAR-T cell-specific toxicities were manageable, consisting of mild to moderate CRS, with no ICANS observed
[102][59].
No clinical data are available regarding efficacy and safety of trispecific constructs in clinical settings, while preclinical data in mice showed strong activity even against cells negative for one of the three targeted antigens
[103][60].
Although available data regarding multispecific CAR-T cells show promise in reducing CD19-negative lymphoma relapse, they are still limited to small phase I trials and some early phase II trials. Moreover, treatment with multispecific CAR-T cell constructs does not have an impact on reducing relapse due to reasons other than antigen escape. To address the latter limitation and avoid T-cell exhaustion, the association of bispecific CAR-T cells with anti PD-1 monoclonal antibodies has been explored, with encouraging early results
[104][61].
5. New Lymphoma Settings
5.1. Hodgkin Lymphoma
Despite the success of frontline therapy, 20–30% of patients with classical Hodgkin lymphoma (cHL) progress or relapse during the course of their disease history. Salvage with high-dose chemotherapy followed by ASCT remains the standard of care in second-line treatment. Targeted therapy with the anti-CD30 antibody-drug conjugate brentuximab vedotin and immunotherapy with anti PD1/PDL1 are currently considered the best options for subsequent relapses and, more recently, their introduction in combination with chemotherapy regimens in the upfront setting further improved outcomes for patients with advanced stage disease
[105,106,107][62][63][64]. Nevertheless, a significant fraction of patients needs alternative approaches.
cHL is characterized by a reduced number of malignant Hodgkin and Reed–Sternberg (HRS) cells and an abundance of inflammatory and immune cells, including reactive T- and B-lymphocytes, macrophages, granulocytes and fibroblast-like cells. CD30, a member of the TNF superfamily, is selectively over-expressed in HRS cells with very low expression in normal tissues, and therefore it is considered a promising target for novel treatments.
5.2. T-Cell Lymphomas
Despite their rapidly expanding use against B-cell malignancies, the development of CAR-T cell therapy for T-cell neoplasms is rather challenging and no products are currently available for this subset of lymphomas.
Most antigens tested as CAR-T cell targets for T-cell neoplasms (such as CD3, CD5, and CD7) are also expressed by normal T-cells
[110][65]. Therefore, the lack of a tumor-specific antigen can cause the eradication of normal T-lymphocytes by CAR-T cell treatment, leading to T-cell aplasia, a possibly life-threatening phenomenon
[111,112][66][67].
Additionally, the expression of their own target antigens by CAR-T cells leads to mutual killing, also known as fratricide, resulting in impaired persistence and efficacy
[113][68].
Although less common than in ALL, circulating tumor cells might be found in peripheral blood. Therefore, T-cell collection could be contaminated with malignant cells, which can be erroneously transduced, as reported in a patient who relapsed by expressing anti-CD19 CAR-tumor cells
[114][69].
One of the fundamental challenges in developing effective CAR-T cell therapies for T-cell lymphomas lies in the identification of suitable target antigens. The ideal target should be specific for the malignant cells and not expressed on normal cells, but T-cell lymphomas originate from mature T-cells and lack a clear distinctive cell surface marker.
6. ‘Off-the-Shelf’ Products for NHLs: Allogeneic CAR-T and CAR-NK Cells
6.1. Allogeneic CAR-T Cells
In contrast to autologous CAR-T cells, off-the-shelf products offer several potential advantages. First, T-cells are collected from healthy donors, thus avoiding the potential detrimental effects of cancer or and cytotoxic agents. Additionally, large quantities of allo-CAR-Ts can be derived from a single donor, enabling the creation of readily available batches of preserved CAR-T cells for immediate patient access. The increased availability of the product may result in a reduced need for bridging chemotherapy and lower costs
[128,129][70][71]. Finally, since allo-CAR-Ts can be created from T-cell subsets that may confer properties such as memory or stemness, a better persistence might be obtained
[127,130][72][73].
Despite the potential benefits of allo-CAR-Ts over autologous CAR-T cell products, if immune cells are sourced from MHC-mismatched donors, these products may lead to graft rejection and to the development of graft-versus-host disease (GVHD). Thus, new sources of T-cells for allogeneic approaches have been explored both in preclinical and clinical studies, including virus-specific T-cells, genetically modified conventional T-cells, and non-conventional T-cells
[127,131][72][74].
The adoption of virus-specific T-cells represents a promising way for mitigating the GVHD risk, given their established role in treating post-transplant viral infections
[132][75]. The safety of this approach was demonstrated in a phase I basket trial involving patients with various B-cell malignancies who received CAR virus-specific T-cells, in which severe GVHD did not occur
[133][76].
T-cell genetic modification offers another strategy to address GVHD and rejection concerns by removing endogenous molecules such as αβ T-cell receptors (TCR) and MHC. In an early clinical study, two infants with R/R ALL were successfully treated with universal CAR-T cells, a product in which CD52 and αβ TCR were disrupted through a transcription activator-like effector nuclease (TALEN) gene editing technique
[134][77]. While TCR suppression mitigated the GHVD risk, genetic disruption of CD52 expression allowed the adoption of the anti-CD52 alemtuzumab monoclonal antibody as part of the lymphodepletion without affecting CAR-T cell activity. More recently, a CRISPR/Cas9 base-edited anti-CD7 CAR-T product characterized by a triple CD52/CD7/βTCR gene suppression pattern showed efficacy against R/R T-ALL
[135][78]. In this case, the gene inactivation of CD52, CD7 and the β chain of the αβ TCR favored the evasion of lymphodepleting therapy, fratricide and GVHD, respectively.
6.2. CAR-NK Cells
Natural killer (NK) cells and macrophages are emerging as highly promising candidates for the development of next-generation off-the-shelf CARs, thanks to their advantageous characteristics. NK cells and macrophages are innate immune system components capable of directly recognizing target cells independent of MHC. Importantly, they do not trigger GVHD and their ability to recognize tumor cells even when MHC molecules are downregulated might prevent antigen escape.
The efficacy of CAR-NK cells has been demonstrated in preclinical models across a range of hematological and solid tumors
[138][79]. Furthermore, in a phase I/II clinical study involving 11 patients with CD19-positive hematological malignancies, promising antitumor effects without significant toxicities were reported following allogeneic UCB-derived CAR-NK cell administration
[63][80].
6.3. Clinical Experiences with NHLs
The first clinical trial of allo-CAR-Ts (phase I ALPHA study, NCT03939026) showing the safety and feasibility of this approach included patients with R/R LBCL or FL after at least two lines of therapy. These patients received a single infusion of healthy donor-derived CAR-T cells without prior lymphodepletion chemotherapy. The ALLO-501 CAR-T cells are genetically modified anti-CD19 CAR-T cells with disrupted TCR alpha and CD52 genes, thus reducing the risk of GVHD and allowing the use of a humanized anti-CD52 mAb (ALLO-647) for selective and transitory host lymphodepletion. Forty-six out of forty-seven enrolled patients received ALLO-501, and the treatment was initiated rapidly, with a median time of 5 days from enrollment to therapy start. No GVHD was reported and limited ICANS and CRS were observed. Cytopenias occurred in 82.6% of the patients, and grade ≥ 3 infections were observed in 23.9% of the cases, similar to what is observed with autologous CAR-T cells. The 6-month CR rate for LBCL patients was 36.4%
[57][81]. The ongoing ALPHA2 study (NCT04416984) is evaluating ALLO-501A (the next-generation product based on ALLO-501 results and lacking the rituximab recognition domains) with either a single or an additional consolidative dose of treatment. In this latter group, patients with ≥SD at day 28 received consolidation with a second ALLO-647 and ALLO-501A cell infusion.
7. Other Strategies for CAR-T Cell Manufacturing Improvement
The vector choice for T-cell transduction can also impact CAR activity and clinical outcomes: traditional “always-on” promoters, such as those carried by retroviral vectors, more commonly lead to tonic signaling and overstimulation of CAR-T cells, which can lead to exhaustion and disease relapse. Enhancing immune cell fitness and limiting or interrupting the interaction between CAR-T cells and target antigens are promising strategies to overcome this limitation
[142,143,144][82][83][84].
Both these objectives can be reached thanks to the use of next-generation gene modification strategies, such as the use of lentiviral vectors, which do not rely on cell division and have a higher transduction efficiency, or with nonviral methods, such as CRISPR/Cas9 technology. Disruption of the regulation of DNA methylation in CAR-T cells has resulted in enhanced T-cell proliferation and a more prolonged antitumor response, as was first observed in a clinical trial as a result of casual disruption via lentiviral integration of the gene TET2. This observation was later confirmed in a preclinical model by knocking out the gene DNMT3A
[145,146][85][86]. The downregulation of the activity of specific genes or the overexpression of target transcription factors can lead to a substantial improvement in T-cell potency, expansion and persistency, and many potential targets are being explored
[147][87].
8. Conclusions
In the last decade, anti-CD19 CAR-T cells reshaped the treatment paradigm for patients with R/R LBCL and MCL, and many steps forward have been made since the first approval in these settings. Axi-cel and liso-cel have already moved to the second line for refractory and early relapsed patients with LBCL and, more recently, CAR-T cell access has been extended to R/R FL. Several reports demonstrated the efficacy and safety of CAR-T cells in patients with CNS lymphoma and could hopefully support the extension of the regulatory approval to this setting. The use of axi-cel as part of frontline treatment for patients with adverse risk LBCL showed promising results in the phase II single arm ZUMA-12 trial, and the ongoing phase III randomized-controlled ZUMA-23 trial will prove whether this approach will be the new SOC for selected patients with very-high-risk LBCL. New strategies are actively being tested to improve CAR-T cell efficacy and accessibility. Dual targeting CD22/CD19- and CD20/CD19-CAR-T cell products showed significant activity against R/R B-NHLs, emerging as a promising strategy to overcome tumor antigen escape mechanisms. Allogeneic CAR products represent an attractive alternative to autologous CAR-T cells thanks to their ‘off-the-shelf’ nature and potential increased antitumor activity, but further data are necessary to thoroughly assess the long-term implications. Finally, early studies demonstrated promising efficacy of CAR-T cells against new disease subtypes, including R/R cHL and, despite the challenges, T-cell lymphoma.