Response to COVID-19 vaccination imaged by PD-L1 PET scanning

Background During a phase 0 clinical trial of an investigational programmed cell death ligand-1 (PD-L1) PET tracer in patients with non-small cell lung cancer (NSCLC), three patients received booster doses of COVID-19 vaccines before PD-L1 imaging. Methods Five patients underwent whole-body PET/CT imaging with a novel PD-L1 tracer, constructed by attaching 89Zr to the anti PD-L1 antibody durvalumab. Intramuscular (deltoid) booster doses of mRNA BNT162b2 COVID-19 mRNA vaccine were coincidentally given to three patients in the month before PD-L1 tracer injection. Results Two recently-vaccinated patients, in remission of NSCLC and receiving non-immunosuppressive cancer therapies (immunotherapy and tyrosine kinase inhibitor respectively), showed increasing PD-L1 tracer uptake in ipsilateral axillary lymph nodes. No asymmetric nodal uptake was seen in a third recently-vaccinated patient who was receiving immunosuppressive chemotherapy, or in two patients not recently-vaccinated. Conclusion Immune response to mRNA BNT162b2 vaccination may involve regulation by PD-L1 positive immune cells in local draining lymph nodes in immunocompetent patients. Trial Registration This trial was registered with the Australian New Zealand Clinical Trials Registry. Registration number ACTRN12621000171819. Date of Trial Registration 8/2/2021. Date of enrolment of 1st patient 11/4/2021. URL of trial registry record: https://www.australianclinicaltrials.gov.au/anzctr/trial/ACTRN12621000171819.


Introduction
There is increasing interest in studying immune responses to cancer and investigating the mechanisms of responsiveness and resistance to immunotherapy by imaging the biodistribution of biomolecules labelled with positron-emitting radioisotopes.These tracers may also provide information on other immune-related processes, including responses to infection and vaccination.Metabolic imaging with 18 F-fluorodeoxyglucose (FDG)-PET is widely used to image non-small cell lung cancer (NSCLC).In recent years, specific PET tracers have also been developed to image components of the immune system, to explore sensitivity and resistance to immune checkpoint inhibitors (ICIs) in cancer patients (Hegi-Johnson et al. 2022a, b).The programmed cell death ligand-1 (PD-L1) is commonly-expressed on cancer cells and on immune cells, including macrophages, dendritic cells and T lymphocytes.PD-L1 is a target for cancer immunotherapy and patients treated with ICIs for NSCLC have shown dramatic improvements in survival (Socinski et al. 2018;Antonia et al. 2018).The programmed cell death protein1 (PD1)/PD-L1 axis is also involved in the immune response to viral infections (Schonrich and Raftery 2019;Wykes and Lewin 2018) but its role in vaccination is unknown.
We report results of imaging of the vaccination response for patients who received a novel PET tracer targeted at PD-L1 (Wichmann et al. 2023) during a cancer imaging trial (Hegi-Johnson et al. 2022a, b), conducted during the COVID-19 pandemic.Coincidentally, three of five patients underwent PD-L1 PET imaging after the mRNA BNT162b2 COVID-19 vaccine (Comirnaty, Pfizer/ BioNTech) was administered by deltoid intramuscular injection.

Materials and methods
The phase 0 ImmunoPET trial investigated safety and biodistribution of a novel PET tracer intended for imaging tumor PD-L1 in cancer patients.The ethics committeeapproved trial was registered with the Australian and New Zealand Clinical Trials Registry (Number 12,621,000,171,819) and all patients gave written informed consent.The tracer comprises a monoclonal antibody to PD-L1, used in cancer treatment, (durvalumab, Astra Zeneca (AZ), Wilmington, DE) bound to the positron-emitting isotope 89 Zr (half-life 78 h) (Wichmann et al. 2023;Rudd et al. 2016).An intravenous injection of 70 megabecquerels (MBq) of 89 Zr-labelled durvalumab is followed by serial PET/CT imaging in the following 5-6 days, allowing time for target binding and elimination of unbound tracer.Eligible patients had advanced (stage IV) PD-L1 positive IV NSCLC but were not required to have active tumor at the time of imaging.

Results
Five patients were recruited (Hegi-Johnson et al. 2023), of whom three coincidentally received COVID-19 mRNA vaccines in the month before tracer administration.PD-L1 imaging results in relation to vaccination are described below by trial identifier number and summarized in Table 1.Table 2 shows changes in maximum standardized uptake value (SUVmax) in right and left axillary nodes at each time point for all patients.
Patient 3 A female aged 58, diagnosed with stage IV lung adenocarcinoma, had nodal and pulmonary metastases on baseline FDG-PET staging in January 2021.An EGFR mutation conferred sensitivity to tyrosine kinase inhibitor (TKI) therapy and she commenced osimertinib (AZ) on January 18th 2021.FDG-PET on August 31st 2021 showed a complete response.Mildly FDG-avid reactive left axillary nodes were reactive to ChAdOx1 nCoV-19 vaccination (Vaxzevria, CSL, Australia), in the left deltoid muscle on July 7th 2021.A second ChAdOx1 nCoV-19 vaccination occurred on September 14th 2021.A third, left deltoid vaccination, BNT162b2 (comirnaty, Pfizer/BioNTech) occurred on February 2nd 2022.Follow-up FDG-PET scan on February 25th 2022 showed remission of NSCLC and FDG-avid left axillary lymphadenopathy, consistent with an inflammatory response to vaccination (Fig. 1a).She enrolled in the ImmunoPET study and received 89 Zr-durvalumab on March 1st 2022, 27 days post vaccination.Serial PET/CT scans recorded the evolving biodistribution of the PD-L1 tracer.No accumulation of the PD-L1 tracer was observed previous tumor sites, consistent with ongoing remission.However, steadily increasing tracer uptake, characterised visually (Fig. 1b), and by sequential SUVmax (Table 2), peaking on the day five scan (Fig. 1b), occurred in the left axillary lymph nodes.Increasing PD-L1 tracer uptake was also observed in the contralateral axilla and bilateral groins, although less intense than in the left axilla (day 5 SUVmax 12.9 left axilla, 6.0 right axilla).
Patient 5 A female ex-smoker, age 63, had a large lung tumor with extensive nodal involvement was diagnosed with Stage IV adenocarcinoma of the lung in November 2019.Following radiotherapy in November 2019, she commenced ICI therapy with pembrolizumab (Merck, Rahway, NJ) in December 2019.She entered a sustained complete remission, documented by serial FDG-PET scans.She received two doses of ChAdOx1n CoV-19 vaccine in 2021.On June 5th 2022 she received a booster with the BNT162b2 COVID-19 vaccine into the left deltoid.On June 6th 2022 an FDG-PET scan showed no

Yes
Both patients with asymmetric axillary nodal uptake showed steadily increasing uptake in left axillary nodes over time, highlighted in Bold, on the same side as recent COVID-19 vaccination.Other patients did not show this pattern evidence of lung cancer but mild FDG-uptake at the deltoid vaccine injection site.She entered the ImmunoPET trial and received the PD-L1 tracer on June 8th 2022, 3 days after vaccination.On serial PD-L1 PET scans, increasing tracer uptake was observed in the left axillary nodes, peaking on day 5, characterised visually (Fig. 2), and by sequential SUVmax (Table 2).No other pathological uptake was observed.The remaining trial patients showed neither significant asymmetrical uptake of PD-L1 tracer in axillary nodes nor steadily increasing SUVmax, as follows; Patient 2 A 46 years old male, diagnosed with stage IV adenocarcinoma of the lung in July 2020, received chemotherapy with carboplatin and pemetrexed, then maintenance pemetrexed and pembrolizumab.On October 18th 2021 he received adrenal radiotherapy for oligo-progressive disease and was enrolled in the ImmunoPET study.He received a second deltoid dose of Comirnaty vaccine on September 26 2021.Carboplatin 500 mg and pemetrexed 500 mg were given on October 4th 2021.PD-L1 tracer injection was on October 19th 2021, 22 days after vaccination.PD-L1 PET scans showed uptake in the adrenal tumor, without significant axillary uptake despite recent vaccination.

Discussion
We observed ipsilateral axillary PD-L1 tracer uptake after vaccination in two patients who were in remission of their lung cancers and were receiving non-immunosuppressive therapies, (osimertinib Ramalingam et al. 2020) and the ICI pembrolizumab respectively (Pardoll 2012).Steadily-increasing SUVmax was measured in nodes draining the site of recent deltoid injections of COVID-19 mRNA vaccines.A third, recently-vaccinated, patient in this study, who showed no significant asymmetric axillary uptake of PD-L1 tracer, had active cancer and received immunosuppressive chemotherapy around the PET-tracer uptake in draining lymph nodes after vaccination has not previously been reported with PD-L1 PET imaging.However, nodal uptake has been reported with FDG-PET (glucose), 68 Ga-DOTATATE PET (somatostatin receptor), 68 Ga-prostate-specific membrane antigen (PSMA, folic acid metabolism) (Orevi et al. 2022) and 11 C-Choline (choline transporter-like protein-1) PET imaging (Schroeder et al. 2021) after COVID-19 vaccinations.Vaccine-related nodal PD-L1 activity detected on PET is potentially much more informative than non-specific nodal uptake due to glucose uptake or folic acid metabolism.PD-L1 is expressed on specific immune-related mononuclear cells, including dendritic cells, B-cells and NK cells (Wu et al. 2019).Germinal centre B cells are important for the generation of high-quality spike protein-specific antibodies following COVID-19 mRNA vaccination, and vaccination has been associated with increased intranodal and circulating T follicular helper cells that express high levels of PD-1 (Locci et al. 2013;Lederer et al. 2022).A compensatory upregulation of PD-L1 in lymphoid tissue to limit this immune response may be expected but has not been previously measured in people.Without correlative biopsies, it would be impossible to know with certainty whether the axillary uptake that we observed after vaccination is a specific finding related to immunity.However, the increasing SUVmax over time in both of our cases is more consistent with specific binding of tracer to PD-L1 positive cells in axillary nodes than with non-specific change due to blood flow or other physiological factors.
The differing vaccine-related findings in our cohort suggest that PD-L1 plays a role in the vaccine response.Both recently-vaccinated patients with cancers in remission and without a history of immunosuppressive therapy had asymmetric and increasing axillary uptake of the PD-L1 tracer suggesting that their imaging results might represent a normal vaccination response.In contrast, the lack of asymmetric axillary uptake in the recently-vaccinated patient with active metastatic cancer who was receiving cytotoxic chemotherapy may represent a subnormal immune response.PD-L1 PET imaging is in its infancy and physiological variations due to processes other than cancer have not been comprehensively described.When unexplained asymmetric and increasing axillary nodal uptake is observed on PD-L1 PET scans when no tumour is present, consideration should be given to vaccination as a potential cause.
New generations of PET tracers are being developed to explore the immune microenvironment of cancer.However, immune PET imaging could be repurposed to illuminate key cellular components of the total body immune response following vaccination.The ability of whole-body PET imaging to identify the anatomic location of processes involved in the immune response and to portray the evolution of the immune microenvironment over time could increase our understanding of the biology of vaccination and produce better vaccines.

Table 1
PD-L1 PET imaging in relation to vaccine booster dose administration

Table 2
Sequential SUVmax in right and left axillary nodes over time