Risk Factors for Pelvic Insufficiency Fractures in Locally Advanced Cervical Cancer Following Intensity Modulated Radiation Therapy

Purpose: To investigate the incidence of and risk factors for pelvic insufficiency fracture (PIF) after definitive chemoradiation therapy for locally advanced cervical cancer (LACC). Methods and Materials: We analyzed 101 patients with LACC treated from 2008-2014. Patients received weekly cisplatin and underwent external beam radiation therapy with 45 Gy in 25 fractions (node-negative patients) or 50 Gy in 25 fractions with a simultaneous integrated boost of 60 Gy in 30 fractions (node-positive patients). Pulsed dose rate magnetic resonance imaging guided adaptive brachytherapy was given in addition. Follow-up magnetic resonance imaging was performed routinely at 3 and 12 months after the end of treatment or based on clinical indication. PIF was defined as a fracture line with or without sclerotic changes in the pelvic bones. D_50% and V_55Gy were calculated for the os sacrum and jointly for the os ileum and pubis. Patient- and treatment-related factors including dose were analyzed for correlation with PIF. Results: The median follow-up period was 25 months. The median age was 50 years. In 20 patients (20%), a median of 2 PIFs (range, 1-3 PIFs) were diagnosed; half were asymptomatic. The majority of the fractures were located in the sacrum (77%). Age was a significant risk factor (P<.001), and the incidence of PIF was 4% and 37% in patients aged ≤50 years and patients aged >50 years, respectively. Sacrum D_50% was a significant risk factor in patients aged >50 years (P=.04), whereas V_55Gy of the sacrum and V_55Gy of the pelvic bones were insignificant (P=.33 and P=.18, respectively). A dose-effect curve for sacrum D_50% in patients aged >50 years showed that reduction of sacrum D_50% from 40 Gy_EQD2 to 35 Gy_EQD2 reduces PIF risk from 45% to 22%. Conclusions: PIF is common after treatment of LACC and is mainly seen in patients aged >50 years. Our data indicate that PIFs are not related to lymph node boosts but rather to dose and volume associated with irradiation of the elective pelvic target. Reducing the prescribed elective dose from 50 to 45 Gy may reduce the risk of PIF considerably.