Diffusion-weighted MRI (DWI) for assessment of response to high-dose-rate CT-guided brachytherapy (HDR-BT) of hepatocellular carcinoma

Abstract

Background

High-dose-rate computed tomography (CT)-guided brachytherapy (HDR-BT) has shown promising results in patients with hepatocellular carcinoma (HCC). While growing evidence shows clear limitations of mRECIST, diffusion-weighted imaging (DWI) has relevant potential in improving the response assessment.

Purpose

To assess whether DWI allows evaluation of short- and long-term tumor response in patients with HCC after HDR-BT.

Material and Methods

A total of 22 patients with 11 non-responding HCCs (NR-HCC; local tumor recurrence within two years) and 24 responding HCCs (R-HCC; follow-up at least two years) were included in this retrospective bi-center study. HCCs were treated with HDR-BT and patients underwent pre- and post-interventional magnetic resonance imaging (MRI). Analyses of DWI were evaluated and compared between pre-interventional MRI, 1.follow-up after 3 months and 2.follow-up at the time of the local tumor recurrence (in NR-HCC) or after 12 months (in R-HCC).

Results

ADC_mean of R-HCC increased significantly after HDR-BT on the first and second follow-up (ADC_mean: 0.87 ± 0.18 × 10⁻³mm²/s [pre-interventional]: 1.14 ± 0.23 × 10⁻³mm²/s [1. post-interventional]; 1.42 ± 0.32 × 10⁻³mm²/s [2. post-interventional]; P < 0.001). ADC_mean of NR-HCC did not show a significant increase from pre-intervention to 1. post-interventional MRI (ADC_mean: 0.85 ± 0.24 × 10⁻³mm²/s and 1.00 ± 0.30 × 10⁻³mm²/s, respectively; P = 0.131). ADC_mean increase was significant between pre-intervention and 2. follow-up (ADC_mean: 1.03 ± 0.19 × 10⁻³mm²/s; P = 0.018). There was no significant increase of ADC_mean between the first and second follow-up. There was, however, a significant increase of ADC_min after 12 months (ADC_min: 0.87 ± 0.29 × 10⁻³mm²/s) compared to pre-interventional MRI and first follow-up (P < 0.005) only in R-HCC.

Conclusion

The tumor response after CT-guided HDR-BT was associated with a significantly higher increase in ADC_mean and ADC_min in short- and long-term follow-up.

Keywords

Diffusion-weighted imaging hepatocellular carcinoma image-guided interstitial brachytherapy therapy response / prediction local ablative therapy

Introduction

Image-guided interstitial high-dose-rate computed tomography (CT)-guided brachytherapy (HDR-BT), a percutaneous ablation technique, was first introduced into clinical practice by Ricke et al. in 2002 (1,2) and has shown promising results with regards to safety, efficiency, local tumor control, and overall survival in patients with hepatocellular carcinoma (HCC) (3,4).

The criteria for an imaging response are usually based on anatomic changes such as size and vascularization. Currently, the modified RECIST (mRECIST) represents the gold standard for the evaluation of a tumor response after treatment for HCC (5,6). However, growing evidence shows clear limitations of mRECIST, since post-treatment contrast enhancement, even several months after therapy, is not the only unique feature of a viable tumor and may also appear in benign tissue due to inflammation and/or granulation (7 –10).

Diffusion-weighted imaging (DWI) reflects motion of free water molecules and allows qualitative and quantitative (on apparent diffusion coefficient [ADC] map) evaluation of changes in tissue cellularity (11 –13). In oncologic imaging, DWI is routinely used for the detection and characterization of liver lesions since benign lesions tend to show higher ADC values than malignant lesions (14). In addition, DWI seems to be useful in determining tumor grading and histopathological parameters that are associated with ADC values. For example, Surov et al. showed that ADC predicts a non-invasive tumor grade and microvascular invasion (15). Similarly, Huang et al. showed that ADC values are correlated with both Ki-67 and HIF-α expression in HCC (16).

Presumably, ADC may also predict a tumor response. In fact, previous studies have shown the ability of DWI to assess a tumor response after Yttrium-90 radioembolization as well as after transarterial chemoembolization (TACE) in patients with HCC (17,18). Kokabi et al. demonstrated that tumors with an objective response by mRECIST had significantly increased ADC value when compared to “non-responders” following Y-90 therapy in patients with infiltrative HCC with portal vein thrombosis (17). Finally, Drewes et al. showed that ADC can predict HCC treatment response after TACE (19). For the evaluation of therapy response after HDR-BT, Wybranski et al. also showed that DWI is a promising imaging marker for early prediction of tumor response in patients with colorectal liver metastases (20).

However, we believe there are no studies that have investigated the feasibility of DWI to predict and evaluate a tumor response in patients with HCC after HDR-BT in short- and long-term intervals.

Therefore, the aim of the present study was to assess whether DWI allows for the prediction and evaluation of a treatment response after HDR-BT in patients with HCC.

Material and Methods

Study population

This retrospective study was approved by the local research ethics committee. After sample size planning, patients with HCC who were treated with CT-guided HDR-BT with an iridium-192 source between 2014 and 2018 that underwent pre- and post-interventional MRIs were included based on a retrospective search in the radiology patient database of two centers.

In center 1, all patients treated between 2017 and 2018 were evaluated. In center 2, we recruited the remaining patients according to a sample size calculation. Treatments in center 2 were performed between 2014 and 2018 (a longer recruitment period due to the longer existing therapy offer for HDR-BT). The patients recruited in center 2 were also part of the previously published study (21). However, the two studies address different questions and aims; the already published study analyzes the prognostic and predictive value of baseline DWI in HCC before HDR-BT with regard to local tumor control and overall survival. In this study, on the other hand, it is examined whether the DWI is suitable for assessing tumor response and whether local tumor recurrence can be detected with this sequence.

Further inclusion criteria signifies that all patients had received a baseline MRI including DWI before brachytherapy, a follow-up MRI including DWI after three months, and a second follow-up MRI including DWI at the time of the local tumor recurrence (non-responder group) or a follow-up MRI including DWI after 12 months (responder = control group). The exclusion criteria included severe motion artefacts, complications during intervention, lesion size <1 cm, follow-up of responding lesions <24 months, incomplete MRI protocol, or locoregional ablative therapy before HDR-BT.

CT-guided interstitial HDR-BT

All patients included in this study underwent CT-guided HDR-BT based on consensus in an interdisciplinary tumor conference. HDR-BT was performed in one single session as described elsewhere in detail (3,4). After analgesia and sedation, brachytherapy catheters were positioned inside the tumor volume under CT guidance, followed by a planning CT scan. Subsequent HDR-BT using an afterloading technique was performed using a 192Ir source. After irradiation, the catheters were removed while administering absorbable gelatine sponge (gelfoam®) to seal the puncture tract (3,4). The applied dose was a minimum of 15 Gy tumor enclosing (3,22).

MRI

MRI examinations were performed using a 1.5-T MR system (Magnetom Avanto, Magnetom Aera Siemens Healthcare, Erlangen, Germany or Achieva, Philips Healthcare, Best, The Netherlands) and protocol included:

T2-weighted (T2W) single-shot and turbo-spin echo sequences with and without fat suppression

T1-weighted (T1W) gradient echo (GRE) sequence with fat suppression before and in the arterial, portal-venous, and late venous phase and hepatobiliary phase after a delay of 20 min after intravenous contrast agent injection (Gd-EOB-DTPA; Primovist, Eovist, Bayer HealthCare, Leverkusen, Germany)

DW sequences with b-values of 50, 400, and 800 s/mm² (center 1) and 0 and 500 s/mm² (center 2)

Image analysis

Reference standard

The classification into non-responding (NR)-HCC and responding (R)-HCC was based on a consensus reading of follow-up imaging over a period of two years after therapy, using the mRECIST: NR-HCCs were defined as treated lesions with (i) persisting and over time increasing enhancement and size (at least 20%) or (ii) new nodular arterial enhancement with corresponding washout and defect in the hepatobiliary phase, whereas R-HCCs were defined as lesions (i) without vascularization in the sense of a disappearance of any intratumoral arterial enhancement or (ii) with a decrease of vascularized size of at least 30% in the arterial phase without washout and without hypointensity during the hepatobiliary phase.

Quantitative and qualitative image analysis

All MRI data were reviewed in consensus by two radiologists within each department (center 1: observers with 2 and 14 years of experience in abdominal MRI; and center 2: observers with 3 and 16 years of experience in abdominal MRI) in three separate sessions with a two-week interval between the sessions: (i) pre-interventional MRI; (ii) first post-interventional MRI; and (iii) second post-interventional MRI.

Location and size measurements

The location of each lesion was recorded, and size measurements were performed in all MRIs. In order to measure the exact lesion size without post-radiogenic changes we measured the size of each HCC lesion, in consensus with all other acquired sequences on T1W axial images in the liver-specific phase on the slice with the largest tumor diameter on both MR examinations.

DWI with ADC measurements

Pre-interventional DWI were evaluated whether (i) HCC lesions showed restricted diffusion, (ii) no diffusion restriction / invisible on high b-value images, or (iii) whether there is no diffusion restriction with a T2 shine through.

In consensus with all other acquired sequences, a region of interest (ROI) was manually drawn on the slice with the largest viable tumor extent (including necrotic areas, if present) on DWI. Later, the ROIs were transferred to the same slice of the ADC map to measure intralesional ADC values including minimal (ADC_min) and mean (ADC_mean) ADC values. ADC measurements were repeated for the same lesions on follow-up MRI images. Furthermore, circular ROIs were also placed in areas of normal liver parenchyma on pre- and post-interventional DWI-MR images.

Statistics

The statistical analysis was performed with SPSS Statistics version 22 (IBM Corp., Armonk, NY, USA) and SAS version 9.4 for Windows (SAS Institute, Inc., Cary, NC, USA).

Sample size planning

Based on an assumption of μ1∼470 ADC_mean increase in responders, as well as μ2∼150 ADC_mean increase in non-responders, and a standard deviation of σ∼300 for the absolute intra-individual ADC_mean change, a sample size of N1 = 24 responders and N2 = 11 non-responders would achieve a power of 81.1% to reject the null hypothesis of equal means with a two-sided significance level of α = 0.05 using a two-sample t-test.

For normally distributed data, such as the ADC_mean and ADC_min of target lesions, paired t-tests were used for comparisons between study visits (before vs. after HDR-BT) and two-sample t-tests were used for comparisons between response groups (intra-individual changes in responders vs. non-responders). For non-normally distributed continuous data (such as lesion size), the Mann-Whitney U test and Wilcoxon test were used instead of the two-sample t-test and paired t-test. For categorical data, such as diffusion restriction, McNemar's test was used for comparisons between study visits, and Fisher's exact test or chi-square test was used for comparisons between response groups.

Results

Patients

In total, 22 patients (20 men, 2 women; mean age = 63.6 ± 24.8) with 11 NR-HCC and 24 R-HCC were included (Fig. 1).

Fig. 1.

Consort chart. DWI, diffusion-weighted imaging; HCC, hepatocellular carcinoma; HD-BRT, high-dose-rate brachytherapy; NR-HCC, non-responding HCC; R-HCC, responding HCC.

In center 1, a total of 11 patients with 21 R-HCC and 2 NR-HCC were selected. In center 2, we recruited the remaining patients (n = 11, 3 R-HCC and 9 NR-HCC) according to the sample size calculation.

Time frame between pre- and post-interventional MRI, location, and size of HCC

Baseline MRI including DWI was performed a mean of 14 ± 20 days before HDR-BT (R-HCC: 16 ± 23 days, NR-HCC: 11 ± 13 days) and the first follow-up MRI with a mean of 88 ± 33 days after HDR-BT (R-HCC: 87 ± 30 days, NR-HCC: 90 ± 40 days). A second follow-up MRI showing a non-response was acquired with a mean of 469 ± 156 days after HDR-BT and a second follow-up MRI in responding lesions was performed with a mean of 357 days ± 30 days after HDR-BT.

A total of 13 lesions were located in the right lobe, 10 lesions in the left lobe, and two lesions in both lobes.

The mean of the overall diameter of R-HCC was 2.25 ± 1.48 cm on pre-interventional MR images and 2.00 ± 1.05 cm on first post-interventional images. Compared to pre-interventional MRI, there was a significant decrease in size (mean size: 1.47 ± 0.72 cm) on second post-interventional MRI. The mean of the longest diameter of NR-HCC did not significantly differ between pre-interventional MR images versus the first and second post-interventional images (5.03 ± 3.17 cm, 4.39 ± 2.69 cm, and 4.51 ± 3.07 cm, respectively).

DWI

Qualitative analysis of DWI

Visual diffusion restriction decreased significantly in R-HCC after therapy: on pre-interventional MRI, 58.3% (14/24) of R-HCC had a visual diffusion restriction while on the first follow-up only 8.3% (2/24) were diffusion restricted (P = 0.001); on the second post-interventional MRI, no lesion showed any diffusion restriction. On the other hand, there was no significant change in diffusion restriction in the NR-HCC: on pre-interventional MRI, 81.8% (9/11) of NR-HCC had a visual diffusion restriction, while on the first follow up, 54.5% (6/11) were still diffusion restricted; and at the time of the local tumor recurrence, there was even a renewed increase in diffusion-restricted lesions to 63.6% (7/11).

There was no significant difference in visual diffusion restriction between R-HCC and NR-HCC on pre-interventional MRI.

Quantitative analysis of DWI- ADC measurements

The ADC_mean of R-HCC increased significantly from 0.87 ± 0.18×10⁻³ mm²/s before treatment to 1.14 ± 0.23 × 10⁻³ mm²/s after treatment (P < 0.001) on first follow-up and showed a further significant increase to 1.42 ± 0.32×10⁻³ mm²/s 12 months after therapy (P < 0.001) (Figs. 2 –4).

Fig. 2.

ADC_mean in responding and non-responding HCC on pre-interventional MRI, first and second follow-up after HD-BRT. ADC_mean of R-HCC increased significantly on first postoperative imaging after treatment and showed a further significant increase of ADC_mean 12 months after therapy (P < 0.001). In contrast, there was no significant change of ADC_mean of NR-HCC on the first MRI after treatment (P = 0.81) and ADC_mean of N-HCC on second follow-up remained almost unchanged compared to first follow-up. ADC, apparent diffusion coefficient; HCC, hepatocellular carcinoma; HD-BRT, high-dose-rate brachytherapy; MRI, magnetic resonance imaging; NR-HCC, non-responding HCC; R-HCC, responding HCC.

Fig. 3.

Responding HCC in an 83-year-old man. (a, b) The pre-interventional axial contrast-enhanced T1-weighted image (arterial phase) shows a hyper-vascular HCC (circle) in segment 8 with hypointensity on the hepatobiliary phase. (c, d) The HCC shows restricted diffusion with high signal on axial DWI image b = 800 s/mm² and low signal on ADC map. The pre-interventional ADC_mean of the HCC was 0.62 × 10⁻³ mm²/s. (e) After HDR-BT, the lesion shows remaining arterial rim enhancement due to radiation on first follow-up. (f) On hepatobiliary phase, the lesion shows central necrosis with a peripheral post-radiogenic hypo-intense rim. (g, h) On DWI, the HCC demonstrated hyper-intense signal and predominantly hyper-intense signal on the ADC map indicating less restricted diffusion compared to the pre-interventional image. The ADC_mean value of the HCC increased to 0.98 × 10⁻³ mm²/. In the second follow-up, there is a shrinkage in the size of the HCC. (i, j) Furthermore, arterial enhancement decreased but was still present due to persisting post-radiogenic changes without a new hypointense defect in the hepatobiliary phase. (k, l) On DWI, the lesions show no restricted diffusion with an increasing ADC value of 1.2 × 10⁻³ mm²/s. ADC, apparent diffusion coefficient; DWI, diffusion-weighted imaging; HCC, hepatocellular carcinoma; HD-BRT, high-dose-rate brachytherapy.

Fig. 4.

Responding HCC in a 64-year-old man. (a) HCC (circle) shows pre-interventional arterial enhancement, (b) hypo-intensity on the liver-specific phase, and (c, d) restricted diffusion with an ADC_mean of 0.75 × 10⁻³ mm²/s. (e, f) After HDR-BT, the HCC shows remaining arterial enhancement and with central necrosis and peripheral post-radiogenic rim hypo-intensity during the hepatobiliary phase. (g, h) On DWI, the HCC demonstrated restricted diffusion with an increasing ADC_mean of 1.00 × 10⁻³ mm²/s. (i, j) After 12 months, the lesion decreased in size, showed no arterial enhancement with central necrosis and peripheral post-radiogenic changes in the hepatobiliary phase. (k, l) There was a further increase of ADC value (1.45 × 10⁻³ mm²/s) with high signal on DWI and high signal on ADC map. ADC, apparent diffusion coefficient; DWI, diffusion-weighted imaging; HCC, hepatocellular carcinoma; HD-BRT, high-dose-rate brachytherapy.

There was no significant change of ADC_mean of NR-HCC (from 0.85 ± 0.24 × 10⁻³ mm²/s to 1.00 ± 0.30 × 10⁻³ mm²/s) on the first MRI after treatment (P = 0.81). The ADC_mean of NR-HCC increased significantly between pre-interventional and second post-interventional MRI (P = 0.018); however, the ADC_mean remained almost unchanged at 1.03 ± 0.19 × 10⁻³ mm²/s compared to the first follow-up (Figs. 2 and 5).

Fig. 5.

Non-responding HCC in a 55-year-old woman. (a–d) On pre-interventional MRI, hyper-vascular HCC (circle) shows a hypo-intense pattern on the liver-specific phase and restricted diffusion with an ADC_mean of 0.75 × 10⁻³ mm²/s. (e, f) Three months after HDR-BT, the HCC shows decreased but remaining arterial enhancement and hypo-intensity on the hepatobiliary phase. (g, h) On DWI, the HCC demonstrated visually partial restricted diffusion with an ADC_mean of 1.02 × 10⁻³ mm²/s. (i, j) After 11 months, the lesion increased in size, shows new focal arterial enhancement with a new defect on the hepatobiliary phase (arrow) indicating local tumor recurrence. (k, l) On DWI, there is visually an increasing diffusion restriction with a persistently low ADC value of 0.98 × 10⁻³ mm²/s. ADC, apparent diffusion coefficient; DWI, diffusion-weighted imaging; HCC, hepatocellular carcinoma; HD-BRT, high-dose-rate brachytherapy; MRI, magnetic resonance imaging.

The ADC_min in R-HCC did not increase significantly from 0.49 ± 0.22 × 10⁻³ mm²/s before treatment to 0.60 ± 0.28 × 10⁻³ mm²/s after treatment on the first follow-up, but there was a significant increase of ADC_min after 12 months (ADC_min = 0.87 ± 0.29 × 10⁻³ mm²/s) compared to pre-interventional MRI and the first follow-up (P < 0.005).

In NR-HCC, there was no significant change of ADC_min (from 0.49 ± 0.22 × 10⁻³ mm²/s to 0.66 ± 0.39 × 10⁻³ mm²/s to 0.60 ± 0.18 × 10⁻³ mm²/s) over time.

The ADC_mean of normal liver parenchyma was 1.02 ± 0.17, 0.89 ± 0.13 and 0.93 ± 0.17 × 10⁻³ mm²/s on pre-interventional MRI, and first and second follow-up, respectively. There were no differences between the ADC_mean of normal liver parenchyma between either R-CC and NR-HCC or between the pre-interventional, first post-interventional, and second follow-up. Furthermore, there were no significant differences between the ADC_mean of normal liver parenchyma between the patients from center 1 and those from center 2.

Discussion

The aim of this study was to investigate whether DWI and quantitative ADC allow the prediction and evaluation of HDR-BT of HCC. Our results show that the ADC_mean of R-HCC increased significantly after therapy in short- and long-term intervals although there was no significant increase of the ADC_mean in the non-responder group in the short interval and no significant change between the first and second follow-up after HDR-BT.

Therefore, the ADC_mean values, especially intra-individual change in the ADC_mean, may be a useful adjunct to evaluate the tumor response of HCC treated by HDR-BT.

The early determination of a tumor response is needed to further improve management of the disease regarding planning reintervention or additional treatments. In addition to the usual morphological image criteria of tumor response, such as size and vascularization, DWI also seems to be able to show therapy effects and therapy response. Previous studies of primary and secondary liver tumors have shown an increase in ADC values after a number of different treatment modalities (23 –28). However, currently, to the best of our knowledge, there are no data on ADC values of patients with HCC undergoing HDR-BT in short- and long-term intervals.

In our study, in contrast to NR-HCC, there was a significant increase in the ADC_mean value in R-HCC after three months. Therefore, DWI seems to allow the early prediction of treatment response after HD-BRT. Wybranski et al. showed in their study of 30 patients (with 43 unresectable colorectal metastases) that DWI is a promising imaging biomarker for the prediction of a tumor response and therapy monitoring in patients with colorectal liver metastases after HDR-BT (20). They found that the mean tumor ADC decreased significantly at first early MRI (two days after treatment) and increased significantly after three months. This could be explained by the fact that brachytherapy leads to immediate cell necrosis of the tissue in the immediate vicinity of the brachytherapy catheters. However, the study did not discriminate between responding and non-responding lesions.

With regard to long-term evaluation, we found a significant increase in ADC_mean and ADC_min in R-HCC between first and second follow-up. We assume this is due to increasing necrosis and declining post-radiogenic changes such as endothelial damage, post-interventional hemorrhage, and cell swelling. In contrast, we could not find any significant change in ADC values for the NR-HCC between the first and second follow-up. Instead, we found a marked increase in the visual diffusion restriction at the time of the recurrence. Similarly, Zhao et al. demonstrated in their meta-analysis of 624 patients (with 712 HCC lesions) that DWI performed better in diagnosing residual or recurrent HCC lesions after TACE and hence ADC values may serve as alternatives for further evaluation of residual or recurrent lesions in patients with HCC after TACE (29).

There is controversial discussion in the current literature regarding the role of pre-treatment ADC values in predicting a tumor response. Yong et al. and Tamm et al. demonstrated that pre-treatment ADC_mean values in responding lesions of patients with colorectal and gastric hepatic metastases were significantly lower than those of non-responding lesions, which could be related to necrosis or a dampened immune response (30,31). However, in our study, there were no significant differences in pre-treatment ADC values in different response groups. Nevertheless, we can support the results of the aforementioned studies with our qualitative DWI results.

The present study has some limitations. The response rates to CT-guided brachytherapy have been reported as extending 90% even in tumors exceeding 10 cm (29). Therefore, the number of patients eligible for study inclusion was relatively low. Furthermore, the timing of imaging acquisition before and especially after treatment between the responder and non-responder groups was not entirely similar. It must be noted that MR data acquisition was not standardized between centers. For example, b-values differed and may have affected the results. However, there was no statistical difference of ADC measurements of normal liver parenchyma between the patients in center 1 and those in center 2; the results, therefore, seem to be robust in light of routine MR acquisition with non-standardized DWI at the two centers. In addition, NR-HCC were larger than R-HCC, which may have an effect on ADC measurements, though there were no significant differences between ADC_mean and ADC_min on pre-interventional MRI. However, larger studies are needed to confirm these results. Furthermore, mRECIST is limited by post-interventional changes. In order to avoid possible post-interventional changes that simulate a local recurrence, we chose a long period of one year after the intervention for the final evaluation.

In conclusion, ADC measurements, especially the increase of ADC values over time, could be a highly useful additional non-invasive diagnostic tool alongside “standard” response analysis for the evaluation of treatment response in early and long-term follow-up after CT-guided HDR-BT in patients with HCC.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Homeira Karim

Maximilian Thormann

Alexey Surov

Christine Schmid-Tannwald

References

Ricke

Wust

Stohlmann

, et al. CT-guided interstitial brachytherapy of liver malignancies alone or in combination with thermal ablation: phase I–II results of a novel technique. Int J Radiat Oncol. 2004;58(5):1496–1505.

Ricke

Wust

Wieners

, et al. Liver malignancies: CT-guided interstitial brachytherapy in patients with unfavorable lesions for thermal ablation. J Vasc Interv Radiol. 2004;15:1279–1286.

Mohnike

Wieners

Schwartz

, et al. Computed tomography–guided high-dose-rate brachytherapy in hepatocellular carcinoma: safety, efficacy, and effect on survival. Int J Radiat Oncol 2010;78:172–179.

Bretschneider

Ricke

Gebauer

Streitparth

. Image-guided high-dose-rate brachytherapy of malignancies in various inner organs - technique, indications, and perspectives. J Contemp Brachytherapy 2016;8:251–261.

Lencioni

Llovet

. Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin Liver Dis 2010;30:052–060.

Kim

Han

K-H

, et al. Evolution from WHO to EASL and mRECIST for hepatocellular carcinoma: considerations for tumor response assessment. Expert Rev Gastroenterol Hepatol 2015;9:335–348.

Vouche

Kulik

Atassi

, et al. Radiological-pathological analysis of WHO, RECIST, EASL, mRECIST and DWI: imaging analysis from a prospective randomized trial of Y90 ± sorafenib. Hepatology 2013;58:1655–1666.

Vossen

Buijs

Kamel

. Assessment of tumor response on MR imaging after locoregional therapy. Tech Vasc Interv Radiol 2006;9:125–132.

Ibrahim

Nikolaidis

Miller

, et al. Radiologic findings following Y90 radioembolization for primary liver malignancies. Abdom Imaging 2009;34:566–581.

10.

Ludwig

Camacho

Kokabi

, et al. The role of diffusion-weighted imaging (DWI) in locoregional therapy outcome prediction and response assessment for hepatocellular carcinoma (HCC): the new era of functional imaging biomarkers. Diagnostics (Basel, Switzerland) 2015;5:546–563.

11.

Koh

Collins

. Diffusion-weighted MRI in the body: applications and challenges in oncology. Am J Roentgenol 2007;188:1622–1635.

12.

Le Bihan

Turner

Douek

, et al. Diffusion MR imaging: clinical applications. AJR Am J Roentgenol 1992;159:591–599.

13.

Le Bihan

Breton

Lallemand

, et al. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology 1986;161:401–407.

14.

Galea

Cantisani

Taouli

. Liver lesion detection and characterization: role of diffusion-weighted imaging. J Magn Reson Imaging 2013;37:1260–1276.

15.

Surov

Pech

Omari

, et al. Diffusion-weighted imaging reflects tumor grading and microvascular invasion in hepatocellular carcinoma. Liver Cancer 2021;10:10–24.

16.

Huang

Meng

, et al. Correlations between ADC values and molecular markers of ki-67 and HIF-1α in hepatocellular carcinoma. Eur J Radiol 2015;84:2464–2469.

17.

Kokabi

Camacho

Xing

, et al. Apparent diffusion coefficient quantification as an early imaging biomarker of response and predictor of survival following yttrium-90 radioembolization for unresectable infiltrative hepatocellular carcinoma with portal vein thrombosis. Abdom Imaging 2014;39:969–978.

18.

Sahin

Harman

Cinar

, et al. Evaluation of treatment response of chemoembolization in hepatocellular carcinoma with diffusion-weighted imaging on 3.0-T MR imaging. J Vasc Interv Radiol 2012;23:241–247.

19.

Drewes

Heinze

Pech

, et al. Apparent diffusion coefficient can predict therapy response of hepatocellular carcinoma to transcatheter arterial chemoembolization. Dig Dis 2022;40:596–606.

20.

Wybranski

Zeile

Löwenthal

, et al. Value of diffusion weighted MR imaging as an early surrogate parameter for evaluation of tumor response to high-dose-rate brachytherapy of colorectal liver metastases. Radiat Oncol 2011;6:43.

21.

Thormann

Surov

Pech

, et al. Local ablation of hepatocellular carcinoma by interstitial brachytherapy: prediction of outcome by diffusion-weighted imaging. Acta Radiol 2022;19:2841851221129714. DOI: 10.1177/02841851221129714. Online ahead of print.

22.

Collettini

Schreiber

Schnapauff

, et al. CT-guided high-dose-rate brachytherapy of unresectable hepatocellular carcinoma. Strahlentherapie und Onkol 2015;191:405–412.

23.

Cui

Zhang

Sun

, et al. Apparent diffusion coefficient: potential imaging biomarker for prediction and early detection of response to chemotherapy in hepatic metastases. Radiology 2008;248:894–900.

24.

Dudeck

Zeile

Wybranski

, et al. Early prediction of anticancer effects with diffusion-weighted MR imaging in patients with colorectal liver metastases following selective internal radiotherapy. Eur Radiol 2010;20:2699–2706.

25.

Eccles

Haider

, et al. Change in diffusion weighted MRI during liver cancer radiotherapy: preliminary observations. Acta Oncol (Madr) 2009;48:1034–1043.

26.

Koh

Scurr

Collins

, et al. Predicting response of colorectal hepatic metastasis: value of pretreatment apparent diffusion coefficients. AJR Am J Roentgenol 2007;188:1001–1008.

27.

Marugami

Tanaka

Kitano

, et al. Early detection of therapeutic response to hepatic arterial infusion chemotherapy of liver metastases from colorectal cancer using diffusion-weighted MR imaging. Cardiovasc Intervent Radiol 2009;32:638–646.

28.

Schraml

Schwenzer

Clasen

, et al. Navigator respiratory-triggered diffusion-weighted imaging in the follow-up after hepatic radiofrequency ablation - initial results. J Magn Reson Imaging 2009;29:1308–1316.

29.

Liu

Fan

, et al. Utility of diffusion weighted imaging with the quantitative apparent diffusion coefficient in diagnosing residual or recurrent hepatocellular carcinoma after transarterial chemoembolization: A meta-analysis. Cancer Imaging 2020;20:3. DOI: 10.1186/s40644-019-0282-9.

30.

Cui

Zhang

Sun

, et al. Apparent diffusion coefficient: potential imaging biomarker for prediction and early detection of response to chemotherapy in hepatic metastases. Radiology 2008;248:894–900.

31.

Tam

Collins

Brown

, et al. The role of pre-treatment diffusion-weighted MRI in predicting long-term outcome of colorectal liver metastasis. Br J Radiol 2013;86:20130281.