Skip to main content
Premium Trial:

Request an Annual Quote

ImaginAb Advancing CD8 PET Imaging for Immunotherapy Personalization

Premium

NEW YORK ─ Immuno-oncology-focused imaging firm ImaginAb is pushing forward with a Phase II trial of its antibody-based CD8 T-cell imaging technology, hoping to eventually establish the approach as a tool for personalizing cancer immunotherapy treatment.

The firm's technology, which uses radiologic imaging to track whether CD8-expressing T cells have infiltrated a patient's tumor, is very different than the molecular strategies that have traditionally made up the backbone of precision oncology. But it could help bring the same personalization that has been possible for molecularly targeted drugs to the more complex world of immunotherapeutics.

In brief, ImaginaAb's immunotherapy biomarker approach relies on the creation of what the company calls a minibody, that, in this case, binds specifically to CD8 T cells. "As the name implies, it's a small antibody," the firm's CEO Ian Wilson explained.

Unlike normal antibodies, though, he said, a minibody doesn't interact biologically with human cells, which means it passes benignly, and more quickly, through the body, apart from where it binds to its intended molecular/cellular target.

Where targeted cancer drugs are concerned, precision oncology has evolved in close parallel with genomic testing technology, with DNA tests offering oncologists a new framework in which to treat their patients using drugs targeted toward the specific molecular or genetic features of an individual's disease.

But the emergence of immunotherapy, among other breakthroughs, has challenged this paradigm, with evidence accumulating that there are aspects of cancer risk, progression, and individual susceptibility to these medicines, that depend on much more complex features than just genetics.

Some in the field have been postulating for years that precision medicine for immunotherapies might require multi-pronged, combinatorial testing strategies, sometimes termed an "Immunogram." And evolution in the field since then has held this up, with molecular biomarkers like PD-L1, microsatellite instability, and tumor mutational burden each seeming to capture only some of the response variation seen to this new class of drugs.

As with targeted therapies, oncologists need to be able to identify which patients should receive these drugs. And as combination strategies increasingly emerge, they will need tools to guide which drugs to give to which patients at which times and in what order. But unlike some of the strongest biomarker relationships seen in the targeted therapy space (EGFR mutations conferring sensitivity to EGFR inhibitors, for example) molecular immunotherapy biomarkers haven't had as binary a relationship to patient response.

For example, while high PD-L1 expression is associated with a higher likelihood of response to pembrolizumab (Merck's Keytruda) in lung cancer patients, a patient with low PD-L1 expression still may respond.

Similarly, there is a chance that a patient with low microsatellite instability, or low tumor mutational burden, may benefit from immunotherapy. These shortcomings of molecular tests in this space have left oncologists anxious, especially about denying patients who may be out of other options a chance to try immunotherapy. The decision is particularly challenging given high-profile examples of individuals who have experienced very robust and long-lasting responses.

As researchers have studied the predictive ability of various markers, one helpful finding has been that some genomic or molecular features seem to capture different populations of responders, suggesting that if added together they could yield a higher predictive power.

Interestingly, ImaginaAb's approach, which takes a more zoomed out view, may prove to be a way of recapitulating and combining some of these other predictors.

In an interview last year, The Parker Institute's Theresa LaValee explained that things like MSI, tumor mutational burden, and PD-L1 all represent different aspects of how a cancer becomes "hot" or immunotherapy-responsive. Regardless of how that process happens, the end result must include the presence of T cells in the tumor. And it is this aftermath that ImaginAb's technology measures.

Coiners of the "immunogram" concept have also argued that immunotherapy response rests ultimately on T cells, calling their activity the ultimate "effector mechanism."

In its Phase II trial with the Dana-Farber Cancer Institute, ImaginAB is now at work imaging patients at baseline and after they have received immunotherapy, measuring how CD8-positive T-cell distribution predicts response, as well as how it changes between these two time points and how well that change corresponds to clinical response.

In the company's Phase I study of the 89Zr-IAB22M2C minibody, which was published last October in the Journal of Nuclear Medicine, investigators recruited six patients, one with melanoma, four with lung cancer, and one with hepatocellular carcinoma, all of whom received an injection of the tracker with PET/CT scans at progressive time points.

According to the authors, infusion of the minibodies was "well tolerated," with no side effects observed. Based on the patterns of uptake in the post-administration imaging, it looked like the tracker was preferentially distributed in predicted CD8-positive T cell-rich tissues, including across different metastatic lesions.

LaValee said last year that she viewed the signal-to-noise ratio in that early Phase I data as very encouraging.

According to Wilson, the Phase II follow up will hopefully reiterate the safety data established in the small Phase I trial, and will begin to define more precisely how well CD8-positive T-cell infiltration predicts future response to treatment, as well as how on-treatment changes in these cells correspond to either response or resistance.

Both could have their own clinical utility, he added. Upfront, or pre-treatment use of the minibodies, he said, will likely make the most sense in combination with other biomarkers like PD-L1. Results from the Phase II and future trials will hopefully help the company prove that if a patient is negative for T cells in their tumor, they are very unlikely to respond to immunotherapy. Especially if they are negative for multiple biomarkers — for PD-L1 and TMB and T-cell infiltration, for example — an oncologist could feel more confident that this isn't going to be an effective option for their patient, he added.

For patients who have at least some biomarker evidence for responsiveness, whether PD-L1, or TMB, or a combination, the ImaginAb minibodies, if used after a patient begins treatment, could also provide a measure of whether patients are indeed responding, giving doctors another time point in which to make decisions.

This type of personalization, Wilson said, could become more attractive as new immunotherapy drugs are developed and physicians face choices about whether, and when, to move patients from something like a checkpoint inhibitor to potentially more effective, but also more toxic options.

An on-treatment biomarker scenario is a very different model of precision medicine than predictive testing, and it doesn't necessarily make sense in all aspects of oncology. For therapies that propose to extend life by only a few months compared to standard chemotherapy, for example, doctors really need to know beforehand if their patient stands to benefit or would do just as well, or better, on standard treatment.

But for immunotherapy, the potential for dramatic and long-term responses is such that a model using early on-treatment biomarkers could be attractive.

Interestingly, molecular/genomic biomarkers are also gaining attention for this type of monitoring paradigm: specifically, liquid biopsy tools that detect fragments of tumor DNA in the blood.

Wilson said that ImaginAb's goal is for its technology to find use in clinical oncology practice. But as it works to establish validity in Phase II and beyond, the firm is also doing a lot of work with and for pharma companies, where its minibodies propose to help drug companies more quickly and accurately assess efficacy of new candidates.

"As a biotech company you are always raising money," he said. "And by working with pharma we offset a need to raise as much." But in the end, this work also contributes toward moving the technology into clinical diagnostics, because clinical data from various drug trials will help establish the validity of the company's approach, as it works toward regulatory approval.

Importantly, several pharma companies have also pledged to work together with ImaginAb, and with each other, to make sure that some of the standardization challenges faced with PD-L1 testing, for example, don't repeat themselves in CD8 imaging.

Specifically, Wilson said, AstraZeneca, Pfizer, and Takeda have all joined a pre-competitive consortium to standardize and lock in a positivity cutoff point for CD8-positive T-cell infiltration within ImaginAb's current trial, which could then be applied consistently across different existing and emerging drugs.

Overall, he added the company's technology is being used in trials of eight of the "top 10" immunotherapies.

Although other companies have antibody technologies for targeted imaging of particular cells and tissues, Wilson said that to his knowledge any potential competitors, one of which is a collaboration between Indi Molecular and GE Healthcare, are all still in preclinical development.

As it pushes forward in immunooncology, Wilson added that ImaginAb is also seeing interest in its approach for other disease areas outside cancer. First on its list is developing a CD4 agent, similar to the current CD8 minibodies, which could be useful for predicting sensitivity to a growing menu of drugs in autoimmune diseases like rheumatoid arthritis or in monitoring patients undergoing transplantation.