CD8-Tracer Is Safe, Shows Biodistribution of T Cells in Tumors

Michael S. Gordon, MD

THE CD8-TRACER 89Zr-IAB22M2C, an anti-CD8 radiolabeled minibody, was found to be safe, well-tolerated, and demonstrated whole-body biodistribution of CD8-positive T cells in tumors and reference tissues, according to first-in-human results presented at the 33rd Annual Meeting of the Society for Immunotherapy of Cancer (SITC 2018).

“This zirconium-labeled minibody targeting CD8 [with] PET /CT imaging is safe and well tolerated in the range of doses studied, and demonstrates robust tracer activity up to 7 days post-injection, [with] a high uptake to background signal ratio,” explained lead author Michael S. Gordon, MD, medical director of HonorHealth Research Institute, in a presentation during the meeting.

Tumor infiltration via CD8+ T cells is associated with favorable outcomes to immunotherapy. While biopsies are used to assess tumor infiltration, Gordon explained that one biopsy does not capture the immunologic heterogeneity existing among various tumors in one patient. Non-invasive imaging, they noted, could allow for a deeper analysis of T-cell infiltration and be a positive correlation with patient outcomes. “As checkpoint inhibitor therapy moves into the frontline and earlier setting, the ability to have a noninvasive technology to assess whole body and intratumoral changes in CD8 trafficking or expansion in response to therapy is crucial,” said Gordon.

The advantages of minibody imaging, he said, is that since it is of a smaller size, there is faster clearance from blood; it is linked to a Zirconium-89 label, which has a long half-life and has centralized distribution. This mechanism to identify these CD8+ cells following injection has been preclinically validated.

The phase I, first-in-human, nonrandomized, open-label two-stage study (NCT03107663) assessed by PET scans using the CD8-tracer to detect whole body and tumor CD8 distribution in 15 patients with metastatic solid tumors. The trial was conducted at Memorial Sloan Kettering Cancer Center, HonorHealth Research Institute, and the University of Pennsylvania.

Patients were aged 31-82 years; nine patients were male and six were female. The included tumors were melanoma (n = 8), non—small cell lung cancer (n = 6), and hepatocellular carcinoma (HCC; n = 1). Two patients had no prior treatment, three had discontinued their previous checkpoint inhibitor therapy, and ten patients had been treated with immunotherapy for a duration of 2 weeks to more than 2 years.

In the first stage, six patients received dose-escalated protein doses starting at 0.2 mg-10 mg (with a static radiation dose of 3mCi IAB22M2C) to establish the safety and determine appropriate scanning parameters. In stage two, the dose-expansion phase, nine more patients were scanned to better evaluate the selected doses of 0.5 mg and 1.5 mg. All patients received 3mCi CD8-tracer once intravenously followed by the serial PET scans over 5-7 days.

All patients were observed for treatment-related adverse events ( TRAEs) and evaluated with blood chemistry, cytokine assay, anti-drug antibodies. Semi-quantitative evaluation of CD8-tracer uptake, biodistribution, and radio-dosimetry were performed.

Results showed that there were no TRAEs, cytokine release syndrome, or blood test abnormal- ities observed. Moreover, there was a transient increase in anti-drug antibodies in one patient that resolved. Additionally, the CD8-tracer rapidly cleared from the blood and was accumulated and saturable in CD8+ rich tissue, specifically in the spleen, bone marrow, and lymph nodes. There was no background issue, Gordon said. The tumor uptake was variable and was seen in 10 patients.

Regarding biodistribution and pharmacokinetics, this was favorable between the 0.5- and 1.5-mg protein doses, with an estimated mean effec- tive radiation of 2.4 rem/mCi, which is consistent with other zirconium-labeled antibody or mini-body technologies, Gordon explained.

All dose levels, however, were found to be safe.

“In the 5- and 10-mg [doses], their biodistributions are increasing in the blood pool as you can see with vascular collection, suggesting that we are above the threshold necessary for adequate imaging—and that led us to the range of 0.5-1.5 [doses].”

The primary tracer excretion was hepatobiliary. The most favorable imaging time point was 24 hours, although tumors can be detected as early as 1-2 hours post-infusion at the earliest, providing a possibility of same-day imaging.

“These results highlight the potential of 89Zr-IAB22M2C to address fundamental questions regarding the presence of CD8+ T cells and their role in cancer and other diseases,” Gordon said.

In preliminary data presented at the 2018 ASCO Annual Meeting, patients with melanoma and HCC who had both received checkpoint inhibi- tors, localization of 89Zr-IAB22M2C was reported in lesions as early as 2 hours following infusion and in the majority of lesions by 24 hours of post-infusion. The Standardized Uptake Values max in 6 target lesions—one intramuscular, two liver, three lymph nodes—ranged 5.85-22.8. There was no radiographically apparent uptake in one patient with NSCLC.²

Next steps following these first-in-human findings from SITC 2018 include a phase II study to correlate imaging with synchronous biopsy and test pre- and post-treatment and response correlation, Gordon concluded.

“This holds the potential for translating imaging correlates with clinical decision making,” said Gordon. “How that will work in our paradigm for treatment remains to be determined using this new tool.”

References:

1. Gordon MS, Tsai F, Postow M, et al. Imaging of tumor infiltrating T cells with an anti-CD8 minibody 89Zr-IAB22M2C in advanced solid tumors: a phase I first-in-human study. In: Proceedings from the 2018 SITC Annual Meeting; November 7-11, 2018; Washington, D.C. Abstract 049.
2. Postow MA, Harding JJ, Hellmann MD, et al. Imaging of tumor infiltrating T cells with an anti-CD8 minibody (Mb) 89Zr-IAB22M2C, in advanced solid tumors. J Clin Oncol. 2018;36 (suppl; abstr e24160). abstracts.asco.org/214/ AbstView_214_218407.html.