Proton craniospinal irradiation reduces active bone marrow exposure in comparative planning study

In nine-patient planning comparisons, proton CSI achieved the lowest activity-weighted marrow dose at 10.9 Gy versus 16.2–19.2 Gy with photon techniques.

KEY POINTS

  • This retrospective treatment-planning study from Technical University of Munich compared four craniospinal irradiation techniques in nine adult patients, generating 36 treatment plans. Five patients had received curative-intent irradiation and four palliative treatment for leptomeningeal disease. Plans included conventional 3DCRT, standard helical tomotherapy (HT), modified x-shaped HT, and pencil-beam scanning proton CSI. All were standardized to 36 Gy or 36 Gy(RBE) using identical target definitions.
  • Proton CSI achieved the lowest activity-weighted exposure of proliferating bone marrow. Weighted bone marrow exposure (WBME) mean dose was 10.9 Gy with protons, compared with 17.3 Gy using 3DCRT, 19.2 Gy with standard HT, and 16.2 Gy with x-shaped HT. Conventional unweighted mean marrow doses were 12.0, 16.0, 18.8, and 16.7 Gy, respectively. WBME was calculated using published regional distributions of active hematopoiesis rather than patient-specific functional marrow imaging.
  • Among photon techniques, marrow-sparing performance depended on the chosen metric. 3DCRT provided the lowest unweighted mean marrow dose and reduced the low-dose bath: marrow V5Gy was 46% with 3DCRT versus 73% with standard HT and 64% with x-shaped HT. However, x-shaped HT achieved the lowest activity-weighted mean marrow exposure among photon techniques (16.2 versus 17.3 Gy with 3DCRT), demonstrating that anatomical marrow volume and hematopoietic activity may lead to different plan rankings.
  • Proton planning also reduced dose outside the target volume. Compared with proton CSI, mean dose outside the PTV was 3.10 times higher with 3DCRT, 4.94 times higher with standard HT, and 4.39 times higher with x-shaped HT. Among photon techniques, x-shaped HT provided favorable sparing of several upper aerodigestive organs, with mean doses of 8.6 Gy to the esophagus, 7.6 Gy to the larynx, and 13.1 Gy to the parotids. Individual organ advantages varied by technique, emphasizing that no single photon approach optimized every OAR.
  • Substantial hematologic toxicity was observed after clinically delivered helical tomotherapy. Among the nine patients, five developed grade 3 leukopenia and two developed grade 3 thrombocytopenia, with no reported grade 4–5 hematologic toxicity. Two patients had received chemotherapy within three months before irradiation and two concurrently. Because all patients were clinically treated using standard HT, these observations cannot establish differences in hematologic toxicity between planning techniques or validate WBME as a clinical toxicity predictor.
  • The proton plans were research approximations rather than clinically optimized treatment plans. Proton planning used a simplified pencil-beam scanning module with a single posterior field, a fixed 1-cm spot-size assumption, no robust optimization, and no modeled range shifter. Historical target margins were relatively large, and the cohort was small and diagnostically heterogeneous. These factors limit generalizability to contemporary proton CSI workflows, especially for patients with leptomeningeal disease receiving shorter palliative regimens.
  • The findings are dosimetric and do not demonstrate comparable oncological efficacy or reduced hematologic toxicity. Differences in active marrow dose were not prospectively correlated with blood-count nadirs or treatment interruptions. The previously reported randomized survival benefit of proton CSI over photon involved-field RT also cannot be attributed specifically to marrow sparing, since those strategies irradiated different anatomical volumes.

CLINICAL TAKEAWAY

Proton craniospinal irradiation provided the greatest reduction in both anatomical and activity-weighted bone marrow exposure in this small comparative planning analysis. For centers without proton access, 3DCRT reduced the low-dose marrow bath, while modified helical tomotherapy improved activity-weighted marrow sparing and selected OAR doses. The clinical predictive value of WBME and the relationship between these dosimetric differences and hematologic toxicity require prospective validation.

SOURCE

Advances in Radiation Oncology

Browse more research Suggest a correction