Session Item

Although survival rates for patients with locally advanced cervical cancer (LACC) are improving, local relapse and toxicity still remain major concerns. Despite the current consensus contouring guidelines for intensity modulated radiotherapy (IMRT) for cervical cancer (CC) suggests to include the whole uterus in the clinical target volume (CTV), there is no unequivocal evidence supporting this indication. Pathological analyses of surgical resection specimens suggest that a 10 mm uterine margin around the primary tumour (GTV) would cover all microscopic disease extension more than adequately.  Magnetic resonance imaging (MRI) is considered the elective imaging modality for identifying GTV in CC. In this context, the recent development of hybrid MRI-radiotherapy systems (MRgRT) may represent a significant step-forward in the clinical management of CC, ensuring the most reliable target identification ever.

Patients with LACC undergoing chemoradiotherapy (CRT) were retrospectively selected. Each patient received 45 Gy delivered in 25 daily fractions using the MRIdian system (ViewRay, Mountain View, California, US). Each patient's original plan (OP) was used as planning reference.

A new treatment plan (PI) was obtained by considering the CTV as GTV plus an expansion of  1,2cm within uterus and up to 2 cm caudally in the vagina plus pelvic elective nodes and parametria (Figure 1).

Dose-volume histogram (DVH) was analyzed to determine the difference in dose values. PTV V_95%, V₄₅ for bowel bag, V₁₀ for body and  D_max, D_mean for different OARs (bowel bag, bowel loops, bladder, rectum, femoral heads) were extracted from DVHs for plan comparison. 

21 patients were finally enrolled for the study. OP and PI plans achieved comparable target dose coverage (Average ∆V_95% = -0,07±0,9).

PI plans delivered a lower D_mean to the all OARs considered (figure2). D_mean difference was higher for bowel bag (Average ∆D_mean = -1,07Gy±0,8), for bowel loops (Average ∆D_mean = -1,36Gy±0,99) and bladder (Average ∆D_mean = -1,31Gy±0,93).

Average D_max was better in PI,  however,  D_max variation was less than 1Gy for all OARs.

Doses to bowel bag was reduced using PI (Average ∆ V₄₅= -6,94cc±5,72).

Moreover, low-dose spillage was less in the PI compared to OP (Average ∆ V₁₀Body= -230,23cc±221,25).

The dosimetric comparison carried out in this study suggests that MRgRT for partial irradiation of the uterine cervix is feasible and linked to dose reduction to all OARs. The observed dosimetric advantages could reduce toxicity rates and be advantageous especially in young patients.

However, the real clinical impact of this approach must be investigated prospectively.