Predicting the Abscopal Phenomenon—Letter to the Editor Open Access

We compliment Nelson and colleagues for their detailed and thorough review in Molecular Cancer Therapeutics regarding the histologic subtypes and biomarkers of immune response that are most likely to predict for the occurrence of an abscopal effect (1). The authors define the abscopal effect as “regression of distant nontreated sites following primary irradiation to a specific site”, the likelihood of which tends to increase in the presence of immunotherapy, particularly immune checkpoint inhibitors (ICIs; ref. 2).

The purpose of this letter is to define another type of locally initiated systemic anti-tumor immune response that we have observed first-hand with intralesionally injected oncolytic adenovirotherapy ± immunotherapy instead of radiotherapy. We refer to this action-at-a-distance activity with oncolytic adenoviruses as the “adscopal effect”. Agents of the common cold with a long safety track record, adenoviruses are the main viral vectors used for the treatment of cancer (3).

As one of the first investigators to administer the 827-base pair-deleted and transgene-less adenovirus, ONYX-015, Tony Reid, a coauthor on this letter, observed the regression of untreated metastatic lesions only in a minority of patients (4). In contrast to ONYX-015, AdAPT-001 is a 50-base pair-deleted oncolytic adenovirus armed with a transforming growth factor beta (TGF-β) trap transgene to neutralize the immunosuppressive and profibrotic cytokine, TGF-β, which several tumor types overexpress. One of these tumor types is sarcomas.

In the phase I/II trial called BETA PRIME (NCT04673942), with which we are involved, AdAPT-001 is given as monotherapy (phase I) and in combination with a previously received—and failed—checkpoint inhibitor (phase II; ref. 5). As expected, an MTD was not reached and, to date, no dose-limiting toxicities or related serious adverse events have been observed either alone or in combination with the checkpoint inhibitor. Less expected, perhaps, is the abscopal, or rather the adscopal activity that we have observed with AdAPT-001 (6) where 5 of 27 patients (18.5%) in the completed part of the Phase I study, 4 of them with sarcomas, evidenced definitive shrinkage of distant, noninjected tumors as shown for one of the patients in Fig. 1 below. In the ongoing phase II part of the trial with AdAPT-001 and a checkpoint inhibitor, evidence of an adscopal effect is estimated in 5 of 15 patients treated to date.

Adenoviremia is detected at quantifiable levels from intratumoral delivery of AdAPT-001. The specific reason to use the word “adscopal” and not “vaccinia-scopal” or “herpescopal”, to borrow terms and examples from one of the reviewers of this editorial, is that adenoviruses, unlike these other viruses, “free ride” on the surface of circulating red blood cells and lymphocytes. These cell carriers protect adenoviruses from antiviral neutralizing antibodies for potential transduction of distant metastases (7–9). Hence, through a combination of direct lysis, and antiviral and antitumor immune responses these adenoviruses may adscopally eliminate these metastatic cells.

Finally, we mention that several authors have attempted to debunk the abscopal effect; one editorial by Siewert and colleagues even went so far as to call the abscopal effect an “urban myth” because it is so exceedingly rare (10). However, based on our experiences with ONYX-015 and AdAPT-001, the adscopal effect is unquestionably not a myth and perhaps based on preliminary evidence from the phase I and II AdAPT-001 clinical trials not so rare either.

See the Response, p. 250