Purpose

Radiation-induced damages in the pelvis can lead to hematological toxicity (HT) that can reduce tolerance to chemoradiation, which is the primary treatment for patients with cervical cancer. The purpose of this study was to investigate the radiation-induced early temporal changes of the diffusion and perfusion characteristics of the bone marrow microenvironment using intravoxel incoherent motion (IVIM) diffusion-weighted MRI (DWI) and their relationships with HT.

Methods

Twenty-eight patients with newly diagnosed cervical cancer were prospectively recruited. All patients received whole-pelvis radiotherapy and concurrent chemotherapy (CRT) as the primary treatment. HT was defined according to the NCI Common Terminology Criteria for Adverse Events (NCI-CTCAE, version 4.03) of grade 3 and above, which required either hospitalization or intervention.¹ Patients were divided into two groups: patients who suffered HT (HT group) and patients who did not suffer from HT (non-HT group). Two sequential MRI examinations were performed on 3.0T Achieva TX scanner, Philips Healthcare at pre-treatment (MRI-1) and week-4 of CRT (MRI-2). DWI was acquired using single-shot spin-echo echo-planar imaging in free breathing with background body signal suppression using 13 b-values (0-1000 s/mm²). DWI was modeled by IVIM (bi-exponential analysis) to generate the parametric maps of the true diffusion coefficient (D) and perfusion fraction (f). Regions of interest (10 mm² each) were placed in both the ilia (designated as marrow inside the field of radiation) and both the femoral heads (designated as marrow outside of the field of radiation), the latter were used as the control areas (Figure 1). Data were expressed as mean ± standard deviation. Wilcoxon signed-rank test was used to compare the IVIM parametric changes in these areas at the two imaging time-points. The degrees of changes in the IVIM parameters expressed in percentages were compared between the HT group and non-HT group using Mann-Whitney U test. Statistical significance was assumed at p≤0.050.

Results

There were 10 patients who experienced HT (mean age 55 ± 14 years old) and 18 patients who did not experience HT (mean age 55 ± 13 years old). In HT group, the D of the ilia showed a positive shift following CRT (mean change 24 ± 33%); on the contrary, in the non-HT group, there would either be no change or a mild negative shift of D following CRT (mean change -6 ± 31%) (Figures 2 & 3A). The trends of changes in D between the two groups were distinctively different (Mann-Whitney U test, p=0.001) (Figure 3A). Although the f of the ilia substantially increased in both groups following CRT, there was no difference in the degree of increment of f between the HT and non-HT groups (Mann-Whitney U test, p=0.596) (Figures 3B & 4). In the femoral heads, the IVIM parameters remained stable following CRT (Wilcoxon signed-rank test, p>0.050).

Discussion

The early changes and trends observed by IVIM-derived diffusion and perfusion were likely a reflection of the underlying pathophysiology in the bone marrow microenvironment induced by radiation.² Patients who experienced HT demonstrated a positive trend of D following CRT, suggestive of the acute radiation effect on the marrow where edema and hemorrhage predominate.^3,4 In non-HT group, the lack of change in D could indicate the marrow was less susceptible to radiation-induced damage. The substantial increased in f of the marrow within the radiation field could be explained by the radiation-induced changes in the arteriocapillary network with vascular congestion and sinusoidal dilatation,⁵ but had limited value in determining the outcome of HT.

Conclusion

IVIM maybe a promising technique to monitor the radiation-induced changes in the marrow, specifically D, which may offer information in predicting the likelihood of HT, offering opportunity for treatment modulation that could minimize radiation-induced marrow toxicity and improve tolerance to CRT.