Background and purpose: For a proper evaluation of the relationship between carbon ion beam dose escalation and local control in the 81 patients with 82 lesions of stage I non-small cell lung cancer, we have identified the incidence of in-field recurrence by collating the dose distribution with the CT images.
Patients and methods: Eighteen fractions over 6 weeks for 47 patients (48 lesions) and nine fractions over 3 weeks for 34 patients were applied in the carbon dose escalation method from 59.4 to 95.4 gray equivalents (GyE) by a 10% increment and from 68.4 to 79.2GyE by a 5% increment, respectively. The radiation target consisted of primary tumor. Image analysis of the patients with local recurrence was systematically performed after the treatment by focusing attention on the enhanced thin slice CT images of the primary lesion. By superimposing the dose distribution on the planning CT image and marking the anatomically identified loci of recurrence, it was possible to establish the relationship between the dose distribution and the incipient loci of recurrence and to classify the recurrence patterns from the differences in the recurrence loci.
Results: Local recurrence was found in 19 (23.2%) out of a total of 82 lesions. It is possible to distinguish between three recurrence patterns: Pattern 1 representing marginal recurrence and patterns 2a and 2b, which are both instances of in-field recurrence. In pattern 1, four recurrences take place from a region on the upper tumor margin. In pattern 2a, 13 recurrences take place from the center of the tumor. In pattern 2b, two recurrences take place from the near-center of the tumor.
Conclusions: For the 15 in-field recurrence lesions (patterns 2a and 2b) after excluding the four marginal recurrence lesions (patterns 1), we have established that the local control shows dose-dependence. Based on this, we have determined the optimal therapeutic dose.