KEY POINTS
- RETHO was a prospective multicenter phase II study evaluating radical-intent thoracic reirradiation for recurrent lung cancer overlapping or adjacent to a previously irradiated high-dose region. Of 64 enrolled patients, 60 were included in the final analysis, comprising 51 patients with NSCLC, eight with SCLC, and one with unknown original histology. Eligibility required recurrence within 5 cm of the previous 50-Gy EQD2 isodose, a treatment interval of at least approximately nine months, Karnofsky performance status ≥70, and FEV1 >40% predicted. Approximately 62% had mediastinal or ultracentral disease.
- Both conventionally fractionated and stereotactic reirradiation were permitted, provided that radical treatment doses and cumulative OAR constraints could be respected. Conventional fractionation was used in 38/60 patients (63%), most commonly 50 Gy in 25 fractions or 60 Gy in 30 fractions, while 22 patients (37%) received SBRT, most frequently 52.5 Gy in seven fractions. Twenty-three patients received chemotherapy, including 14 sequentially and nine concurrently. The median interval between radiation courses was 29 months, and median follow-up was 26.5 months.
- Cumulative dose constraints were explicitly incorporated into treatment planning. Protocol-specified limits included maximum cumulative doses of 75 Gy₃ EQD2 to the spinal cord for intervals up to five years, increasing to 87 Gy₃ for longer intervals, 100 Gy₃ to the esophagus, and 110 Gy₃ to the trachea and main bronchi. Additional constraints included 90 Gy₃ for major vessels and brachial plexus and a cumulative mean dose limit of 22 Gy₃ for both lungs minus GTV. The protocol permitted small-volume exceptions for selected structures, and target undercoverage was allowed when necessary to prioritize OAR protection.
- High cumulative tumor doses were delivered despite substantial overlap with previous radiation fields. Among 59 patients with evaluable dosimetry, median reirradiation dose was 60 Gy₁₀ EQD2 (IQR 50–77), and median cumulative maximum PTV dose reached 130 Gy₁₀ EQD2. Median reirradiation PTV coverage was 96%, although 26 patients (44%) had PTV coverage below 95% and 19 (32%) below 90%. Cumulative esophageal Dmax exceeded 80 Gy₃ in 20 patients, while 26 patients received tracheal Dmax above 80 Gy₃. In the subgroup with bronchial Dmax above 100 Gy₃, the median reached 113.3 Gy₃, demonstrating the importance of interpreting maximum point doses alongside permitted small-volume exposure.
- No grade 4–5 acute or late treatment-related toxicity was reported, despite the high cumulative doses and frequent ultracentral disease. Ten patients (16.7%) experienced grade 3 toxicity, including seven patients (11.7%) with grade 3 radiation pneumonitis. No airway or esophageal fistula, stenosis, ulceration, major hemorrhage, or radiation myelopathy was documented. However, three patients did not complete treatment, including one with grade 3 dyspnea and two with other serious clinical complications.
- Pulmonary toxicity was significantly associated with cumulative lung V20 and mean lung dose. Among 28 patients with grade 2–3 pneumonitis, median cumulative lung V20 was 23.4% versus 16.5% in patients with grade 0–1 pneumonitis (p=0.024). Median cumulative mean lung dose was 19.4 versus 15.1 Gy₃ (p=0.021). In univariable logistic regression, cumulative V20 was associated with higher pneumonitis risk (OR 1.16 per percentage-point increase, 95% CI 1.01–1.34; p=0.042), as was cumulative mean lung dose (OR 1.25 per Gy₃, 95% CI 1.02–1.52; p=0.031). Neither cumulative nor reirradiation-course lung V5 showed a significant association.
- Oncological outcomes demonstrated meaningful survival but persistent local failure. Median overall survival was 30.1 months (95% CI 20.3 to not reached), with 56% two-year OS (95% CI 43–72%). Local failure occurred in 25 patients, corresponding to median local control of 28 months and 53% two-year local control (95% CI 40–70%). The study lacked a non-reirradiated comparator, included heterogeneous histologies and treatment regimens, and had limited follow-up for very late toxicity. Furthermore, organ-specific rigid dose accumulation and allowances for small-volume dose exceedances limit direct translation of the reported cumulative limits into universal safety thresholds.
CLINICAL TAKEAWAY
Prospective RETHO data suggest that selected patients with recurrent lung cancer can receive radical-dose thoracic reirradiation with acceptable observed severe toxicity when cumulative OAR exposure is carefully evaluated and target coverage is adapted when necessary. Cumulative lung V20 and mean lung dose were associated with grade 2–3 pneumonitis, emphasizing the importance of pulmonary dose-volume assessment alongside protection of central thoracic structures. The absence of grade 4–5 toxicity in 60 patients does not establish the safety of these cumulative constraints for broader populations or longer follow-up.
SOURCE
International Journal of Radiation Oncology, Biology, Physics