How would you manage a patient with a growing central liver metastasis previously treated with SBRT who is ineligible for chemotherapy?

This is a challenging scenario for which there is no quick/easy solution. My initial assessment would involve obtaining a further understanding of the patient’s performance status/disease course/biology (is this an isolated site of recurrence vs polymetastatic progressive disease), prior SBRT dose and volume, time interval since prior SBRT, determination of in-field vs. marginal vs. out-of-field, and suitability for other local treatment strategies.

A constellation of factors which would push me away from re-SBRT would include prior ablative doses (~BED10 ≥ 100) of SBRT with in-field recurrence within a 6-12 month interval. With these factors, I think the biology has suggested relative radioresistance and the doses required to achieve durable local control with re-SBRT are unlikely to be feasible/safe. Additionally, I would be cautious of ablative doses of re-SBRT for patients with more aggressive/systemic biology with multiple sites of progressive disease for which local control at the re-irradiated site is unlikely to alter the disease trajectory and the risk/benefit profile may not favor aggressive local therapy.

If deemed appropriate for local therapy - I think multidisciplinary collaboration is key for a case like this. Review with surgical and interventional radiology colleagues regarding the feasibility of resection, ablation, or transarterial therapy would be quite helpful. If suitable, these would be preferable.

If there is not a suitable alternative, RT dose/fractionation will be dependent upon dose tolerance to adjacent organs at risk, most notably liver, duodenum/small bowel, and central biliary tree. Unfortunately, there are no high-quality resources to guide re-irradiation dose tolerances for this scenario. Tissue repair capacity and interval for repair is less well studied for these OARs (particularly bowel), and therefore, my practice has generally been to proceed more cautiously assuming little to no repair. The Stanford group (Osmundson et al., PMID 25659885 and Toesca et al., PMID 27865544) has a couple of nice resources for central biliary tree tolerances in the de-novo setting. They suggest V_BED1072 < 21 cc/V_BED1066 < 24 cc and V_BED1040 < 37cc/V_BED1030 < 45cc, respectively. Similarly, we have considerable data regarding duodenum/small bowel dose tolerances, and generally, 5-fraction max doses in the range of 33-35 Gy (perhaps voxel doses as high as 38-40) are safe. The recent HyTEC series can provide guidance on liver constraints ( Miften et al., PMID 29482870). I would suggest reviewing the cumulative plan sum- assuming minimal to no repair would likely be safe. Escalating beyond that would be dependent upon risk/benefit based upon the scenario.