Top Line: We all have the feeling that locally advanced NSCLC could surely benefit from dose escalation.
The Study: The problem is, we haven’t figured out a consistent way to escalate dose without negative effects on toxicity and mortality. PET-BOOST was a randomized phase 2 trial from the Netherlands that sought to determine if PET-based dose escalation could improve freedom from local failure in patients receiving definitive chemoradiation for stage II-III NSCLC. Important inclusion criteria included primary tumor size >4cm and a maximum SUV>5. Another important factor to consider is that the “standard” dose-fractionation scheme wasn’t the 60 Gy in 30 fractions we’re accustomed to–it was 66 Gy in 24 fractions. This was the dose delivered to involved nodal stations. The primary tumor was simultaneously dose-escalated as high as possible (and to at least 72 Gy) while respecting OAR constraints. An important side note is that 49% of eligible patients were excluded from the trial as the minimum boost dose could not be achieved without exceeding normal tissue constraints. Each patient had 2 treatment plans generated: one with a SIB to the entire primary tumor PTV and another with a SIB to the portion of the primary with ≥50% of the maximum SUV. The median PET-GTV was 29cc compared to 99cc for the standard GTV. Both treatment plans were required to have the same mean lung dose. Patients were randomly assigned to treatment with the standard plan or the PET-boost plan. The median PTV dose in the standard arm was 79Gy (77-82Gy), and the median PTV dose in the PET arm was 84Gy (82-91Gy). At 1 year, FFLF was 97% in the standard arm and 91% in the PET arm. At 18 months, FFLF was 89% and 82%, respectively. While <10% had local or regional recurrence, a third in each arm had distant failure. As we can see, the PET-based boost modestly increased dose without improving control. Half of patients in each arm had grade 3+ toxicity (54 v 53%) including 19% v 17% with esophagitis/dysphagia and 19% v 32% with pulmonary events. Furthermore, 14% experienced grade 5 events mostly resulting from hemorrhage, esophageal fistula, and pulmonary toxicity. As the trial progressed, these events were less common after excluding patients with tumor encasing major vessels and using a stricter 5mm esophagus PRV constraint. The goal of PET-BOOST was to select a technique to test in a larger randomized trial. Despite the excellent local control, neither technique was chosen for further investigation due to the high rate of grade 5 toxicity.
TBL: While PET-based dose escalation improved tumor control in the PET-BOOST trial, there was a higher than anticipated rate of grade 5 events resulting from bleeding, esophageal toxicity, and pulmonary toxicity.
- Cooke, Radiother Oncol 2023