Background: AST-008 is a potent spherical nucleic acid configuration of a toll-like receptor (TLR) 9 agonist oligonucleotide being developed for the treatment of MCC and CSCC in patients (pts) progressing on immune checkpoint inhibitor (CPI) monotherapy. Previously presented data suggests AST-008 elicits a Th1-type cytokine response and immune cell activation. This parallel-arm phase II design will permit assessment of efficacy in two advanced skin cancers with high unmet medical need. Methods: The study has an open label, multicenter phase Ib/II dose escalation/expansion design. The phase 1b dose escalation used a 3+3 design with increasing doses of intratumoral (IT) AST-008 plus standard flat dose pembrolizumab. The phase II dose expansion is using the recommended regimen (RP2D) of IT AST-008 plus flat dose pembrolizumab or cemiplimab to treat two cohorts of pts with advanced/metastatic MCC or CSCC, respectively. The key objective of the phase II expansion is to provide an estimate of preliminary efficacy of IT AST-008 with pembrolizumab or cemiplimab in pts whose disease progressed on a single-agent CPI therapy. Endpoints also include safety, pharmacokinetic and pharmacodynamic assessments. Pts must have at least two evaluable lesions by RECIST v1.1. One or more lesions may be injected with AST-008, but one lesion (the “witness” lesion) must remain un-injected throughout the study to differentially assess effects of AST-008 on the injected and witness lesions. The RP2D of AST-008 was derived from the dose escalation phase together with pembrolizumab. That RP2D is being used not only in the MCC cohort where AST-008 is combined with pembrolizumab, but also in the CSCC cohort where AST-008 is combined with cemiplimab. To ensure safety of the AST-008/cemiplimab combination, the first 6 pts in the CSCC cohort will be monitored for dose limiting toxicities through the first 5 weeks of dosing, and then the data review committee will assess safety before approving additional patient enrollment to that cohort. Each expansion cohort will enroll up to 29 pts with a Simon two-stage optimal design. In the first stage, 10 pts will be accrued. If there are 0 responses in these 10 pts, enrollment in that arm will be stopped. Otherwise, 19 additional pts will be accrued for a total of 29. The null hypothesis will be rejected if 4 or more responses are observed in 29 pts. This design yields a type I error rate of 0.05 and power of 80% when the true response rate is 20%. The planned phase II enrollment is 58 pts across about 15 US-based sites. Clinical trial information: NCT03684785.