What SBRT dose and constraints would you use for a primary endobronchial NSCLC that is too thick for brachytherapy?

It really depends on the size and location of the lesion with respect to the proximal bronchial tree and other mediastinal structures (such as the esophagus). SBRT, as commonly understood (e.g., 50 Gy in 5 fractions), strikes me as risky in most such scenarios. Something more like 60 Gy in 15 fractions is more likely what I would recommend.

I agree with Dr. @Dr. First Last that I would consider a small endobronchial primary “ultracentral” and therefore have used 60 Gy in 15 fractions. I have asked the planner for a more homogeneous dose distribution for these cases compared to SBRT. This mod hypofrac approach is likely safer than a 5-fraction SBRT plan and more efficacious than a conventional 30-fraction plan. The LUSTRE publication is a good resource on the efficacy and safety of this regimen. We use the limits from that trial for these cases.

Swaminath et al., PMID 39298144.