To test the efficacy and feasibility of pelvic bone marrow sparing intensity modulated radiotherapy (PBMS-IMRT) in reducing grade 2 or higher hematological toxicity (HT2+) for patients with cervical cancer treated with concurrent chemoradiotherapy.

A total of 164 patients with Stage Ib2–IIIb cervical cancer were prospectively enrolled from March 2018 to March 2019 at a single center and were randomly allocated into the PBMS group or the control group. The control group received weekly cisplatin concurrently with IMRT, followed by intracavitary brachytherapy. The PBMS group additionally received PBM dose constraint. The dosimetric parameters of the pelvic bone (PB) and the subsites including hip bone (HIP) and lumbosacral spine (LSS) and the corresponding bone marrow were recorded. The endpoint of the trial was acute hematologic or gastrointestinal toxicity. Receiver operating characteristic curves were used to derive optimal dosimetric planning constraints.

Eighty-two patients in the PBMS group and 82 in the control group were enrolled for statistical analysis. The incidence of HT2+ in the PBMS group was 50.0%, significantly lower than the 69.5% incidence in the control group (P = 0.02). Patients with PB V40 ≥ 28% were more likely to experience HT2+ (OR = 2.85, P = 0.006), while the incidence of grade 2 or higher gastrointestinal toxicity (GT2+) events did not differ significantly between the two groups (P > 0.05). Dosimetric parameters of LSS showed stronger associations with HT2+ than other subsites. The patients with LSS V10 ≥ 87% and LSS mean ≥ 39 Gy were more likely to experience HT2+ (OR = 3.13, P = 0.001;OR = 3.03, P = 0.002, respectively).

PBMS-IMRT reduced HT compared with IMRT alone. Efforts to maintain LSS V10 < 87%, LSS mean < 39 Gy and PB V40 < 28% simultaneously may reduce the risk of HT2 +.

The trial was registered with Chinese clinical trial registry ( ChiCTR1800015069 ).