Intravoxel incoherent motion MRI in primary rectal cancer: correlation with histologic prognostic factors
Zhe Han1,2, Juan Chen2, Min Chen2, Chen Zhang2, and Lizhi Xie3

1Chinese Academy of Medical Science and Peking Union Medical College, Beijing, China, People's Republic of, 2Department of Radiology, Beijing Hospital, Beijing, China, People's Republic of, 3GE Healthcare, MR Research China, Beijing, China, People's Republic of

Synopsis

In this study we compared the association of intravoxel incoherent motion (IVIM) derived parameters with the histologic grade, N-stage, EGFR expression and K-RAS gene mutation of primary rectal cancer. Significant correlations were found between D values and differentiation grade, D* values and N-stage, f values and N-stage. IVIM derived parameters may be a promising imaging biomarker of tumor aggressiveness and prognosis.

Purpose

Several different models for the signal decay in diffusion weighted imaging (DWI) have been proposed. These include the classical monoexponential model with the apparent diffusion coefficient as free parameter and the quite popular intravoxel incoherent motion (IVIM) model, which includes perfusion effects present at low b-values[1]. In this study we compared the association of IVIM derived parameters with the histologic grade, N-stage, EGFR expression and K-RAS gene mutation of primary rectal cancer.

Material and Methods

This retrospective study was institutional review board approved, and informed consent was waived. 35 patients with rectal adenocarcinoma confirmed by pathology underwent MR imaging before surgery. Research sequences included diffusion-weighted magnetic resonance (MR) imaging with 15 b values (10,20,30,40,60,80,100,150,200,400,800,1000,1200,1500,2000sec/mm2). In all patients, surgery was performed without neoadjuvant therapy. The region-of-interest (ROI) was selected based on the b1000 image, avoiding the area of cystic and necrosis. Pure molecular diffusion (D), perfusion-related diffusion (D*) and perfusion fraction (f) were calculated. The average value of each parameter in different ROIs was recorded. The differences between the histopathologic parameters were assessed by using the Chi-square test, including t-test (for normally distributed data) and Rank sum test (for non-normally distributed data). The histopathologic parameters included A (differentiation grade: A1, well differentiated; A2, moderately differentiated; A3, poorly differentiated), B (N-stage: B1, lymph nodes involvement; B2, lymph nodes without involvement), C (EGFR expression: C1, positive; C2, negative) and D (K-RAS gene mutation: D1, positive; D2, negative).

Results

The data of the D values was non-normally distributed, While the data of both the D* and f values were normally distributed. Mean D values of the well differentiated group (A1) , moderately differentiated group (A2) and poorly differentiated group (A3) were(1.133±0.451)×10-3mm2/s、(0.690±0.373)×10-3mm2/s and(0.519±0.253)×10-3mm2/s respectively. Mean D values of A1 were higher than that of A2 (Z=-2.283, p=0.011, p<0.05 statistically significant) and A3 (Z=-2.417, p=0.0075), whereas there was no significant difference in mean D values between A2 and A3 (Z=-1.195, p=0.119). The ROI region selection referent to Figure.1. Mean D* values of B1 and B2 were(40.523±17.288)×10-3mm2/s and(65.100±12.602)×10-3mm2/s respectively. There was a significant difference in mean D* values between B1 and B2 (t=4.553, p=0.000). Mean f values of B1 and B2 were(0.491±0.172)and(0.385±0.139)respectively. There was a significant difference in mean f values between B1 and B2 (t=-1.835, p=0.042). There were no significant differences in D, D* or f values when stratifying patients according to EGFR expression and K-RAS gene mutation. The rest of the specific data was recorded in Table.1-4.

Conclusion

Significant correlations were found between D values and differentiation grade, D* values and N-stage, f values and N-stage. IVIM–derived parameters may be a promising imaging biomarker of tumor aggressiveness and prognosis.

Acknowledgements

No acknowledgement found.

References

[1] Le Bihan D, Breton E, Lallemand D, Grenier P, Cabanis E, Laval-Jeantet M. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology, 1986, 161(2):401-407.

Figures

Figure.1 Mean D values of well differentiated group (A1, 1.24 ×10-3mm2/s), moderately differentiated group (A2, 0.449 ×10-3mm2/s) and poorly differentiated group (A3, 0.38 ×10-3mm2/s).

Table.1 Mean D values, mean D* values and mean f values of A1, A2 and A3 (A1, well differentiated; A2, moderately differentiated; A3, poorly differentiated). Mean D values of A1 were higher than that of A2 and A3, whereas there was no significant difference in mean D values between A2 and A3; There were no significant difference in mean D* values or mean f values between any two of A1, A2 and A3.

Table.2 Mean D values, mean D*values and mean f values of B1 and B2 (N-stage: B1, lymph nodes involvement; B2, lymph nodes without involvement).

Table.3 Mean D values, mean D*values and mean f values of C1 and C2 (EGFR expression: C1, positive; C2, negative).

Table.4 Mean D values, mean D*values and mean f values of D1 and D2 (D1, positive; D2, negative).



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
1969