Marginal benefit.

Top Line: We’ve explored adaptive replanning for targets and surrounding organs at risk that can see significant interfraction movement, but where do we stand with replanning for tumors that classically diminish while on beam?

The Study: After all, this seems like a real no-brainer use of replanning. Based on this logic, all patients receiving definitive radiation for non-small cell lung cancer (NSCLC) after 2012 have undergone adaptive replanning at a large Dutch hospital. Adaptive replanning a decade ago, you ask? Well, here they mean old school re-sim and replan based on daily imaging results, not actual auto-replanning done on the table. This is a retrospective look at outcomes before (n=184 treated 2010-2012) and after (n=255 treated 2013-2018) implementation. All patients received 50–66 Gy/25–33 fractions with concurrent platinum-based chemo, but radiation techniques were quite drastically different between the two eras. While daily cone-beam CT (CBCT) was performed during the 2010-2012 period, alignment was to vertebrae, not tumor—yes, begging the question of why attain daily 3D-imaging. The gross tumor volume (GTV) was contoured on the mid-ventilation phase on a 4D-sim with a 5 mm clinical target (CTV) margin and the final planning target volume (PTV) comprising a 10 mm radial margin and 18 mm sup/inf margin. In the 2013-2018 period, tumor was contoured to encompass movement on all phases of the 4D-sim to create a more modern iGTV, again with a 5 mm CTV margin but only a 4 mm radial and 5 mm sup/ing PTV margin (7/8 mm on nodes). Again daily CBCT was obtained, but alignment was to tumor and re-sim / re-planning was done at the discretion of the treating physician based on tumor response (it is unclear how many and how far into treatment patients were re-planned). As you now might guess, median planning target volumes (including the re-planned ones) were slashed from 456 → 270 cc between treatment eras. While this clearly led to much lower heart and lung dosimetry, it also resulted in far less pneumonitis and…wait for it…even better overall survival (43% → 56% at 2 years). Before you cry inherent bias due to better modern systemic therapies, this (1) is before immune checkpoint inhibitors were approved for this indication in Denmark in 2019 and (2) can plausibly be attributed to fewer patients with progression of disease (HR 0.80) as well as grade 2 (50% → 20%), grade 3 (21% → 7%), and most importantly grade 5 (6% → 0.4%) pneumonitis.

TBL: Significantly reducing planning margins with 4D-generated targets, daily 3D-image alignment to tumor, and replanning upon tumor response can slash toxicity and maximize survival benefits achieved with definitive radiation for NSCLC. | Møller, Radiother Oncol 2022