KEY POINTS
- This nonrandomized prospective single-institution cohort with a contemporaneous comparator evaluated 361 patients with T4N0–3M0 nasopharyngeal carcinoma and at least 0.1 cm³ overlap between the primary-tumor PTV and temporal-lobe PRV. The modified strategy was used in 104 patients and standard planning in 257.
- Standard planning prescribed 70 Gy uniformly to the primary-tumor PTV. The modified approach divided that PTV into a non-overlapping subvolume receiving 70 Gy and the portion overlapping the temporal-lobe PRV receiving 66 Gy, delivered using a simultaneous integrated boost approach.
- Five-year cumulative radiographic temporal lobe injury fell from 36.2% with standard planning to 23.2% with the overlap-adapted strategy (P=.019). After multivariable adjustment, the modified strategy remained independently associated with lower injury risk, hazard ratio 0.56 (95% CI 0.34–0.94; P=.029).
- In propensity-matched lobe-level analysis involving 205 temporal lobes per group, the modified strategy again showed lower temporal-lobe injury risk, hazard ratio 0.60 (95% CI 0.37–0.95; P=.029). Temporal-lobe Dmax and D1cc were significantly lower with modified planning, both P<.001.
- The reduction in clinically symptomatic temporal-lobe injury was smaller and did not reach statistical significance: five-year incidence was 9.2% versus 13.5% (P=.095). A bilateral PTV–temporal-lobe overlap ≥1.0 cm³ independently identified higher-risk anatomy, with an injury hazard ratio of 1.54 (95% CI 1.04–2.29; P=.033).
- Selective dose reduction did not appear to compromise cancer control. Five-year local relapse-free survival was 92.3% versus 89.6% (P=.577) and overall survival 84.7% versus 84.4% (P=.896); no marginal recurrence occurred solely within the 66-Gy overlap subvolume.
- Treatment allocation reflected evolving clinical practice rather than randomization, and the comparative analyses were therefore vulnerable to residual selection bias despite multivariable adjustment and propensity matching. The strongest endpoint was radiographic rather than symptomatic temporal-lobe injury, making independent prospective validation important.
CLINICAL TAKEAWAY
For T4 nasopharyngeal tumors where adequate PTV coverage conflicts directly with temporal-lobe constraints, selectively prescribing 66 Gy only to the overlap region may improve the therapeutic ratio without sacrificing observed local control. This is clinically influential prospective evidence, but the nonrandomized design prevents treating the strategy as definitively established.