Application of Intravoxel Incoherent Motion in the Evaluation of Hepatocellular Carcinoma after Transarterial Chemoembolization
, Ping Han 1,2, Jie Yu 1,2,* and Qian Li 1,2,*
Round 1
Reviewer 1 Report
This study aims to assess the efficacy of quantitative parameters of intravoxel incoherent motion (IVIM) diffusion-weighted imaging (DWI) for hepatocellular carcinoma (HCC) diagnosis after transarterial chemoembolization (TACE). The authors concluded that quantitative IVIM-DWI was effective for evaluating tumor viability in HCC patients treated with TACE and may be helpful for non-invasive monitoring of tumor viability.
My comments:
1. Although this study only included 50 HCC patients, it showed quite consistent data in these patients.
2. The authors mentioned that many patients in this study did not have pathological results after TACE, so their gold standard for the determination of TAA and TNA region was mainly based on TACE retreatment results and Li-RADS standards. Thus, the determination of TAA and TNA region were all purely by senior technicians. It still could have personal bias. There should have at least a few cases that have received surgical excision after TACE. So the authors could do the mapping to check the correlations between the pathology examination and Li-RADS standards for the judgement of the TAA or TNA regions.
Author Response
Response to Reviewer 1 Comments
This study aims to assess the efficacy of quantitative parameters of intravoxel incoherent motion (IVIM) diffusion-weighted imaging (DWI) for hepatocellular carcinoma (HCC) diagnosis after transarterial chemoembolization (TACE). The authors concluded that quantitative IVIM-DWI was effective for evaluating tumor viability in HCC patients treated with TACE and may be helpful for non-invasive monitoring of tumor viability.
My comments:
Point 1: Although this study only included 50 HCC patients, it showed quite consistent data in these patients.
Response 1: We would like to thank you for the time and effort spent in reviewing the manuscript. We appreciate your positive comments.
Point 2. The authors mentioned that many patients in this study did not have pathological results after TACE, so their gold standard for the determination of TAA and TNA region was mainly based on TACE retreatment results and Li-RADS standards. Thus, the determination of TAA and TNA region were all purely by senior technicians. It still could have personal bias. There should have at least a few cases that have received surgical excision after TACE. So the authors could do the mapping to check the correlations between the pathology examination and Li-RADS standards for the judgement of the TAA or TNA regions.
Response 2: Thank you very much for the comments and suggestions that greatly help make this manuscript much better!
We apologize for misleading the reviewer by misrepresenting. In fact all of the 50 participants in this study had TACE procedure after MR examination without surgical resection and therefore they had no pathological findings.
In another study, which we have not published, eight patients treated with TACE for hepatocellular carcinoma who performed MR underwent surgical resection within one month and obtained pathological results. Of these 8 patients, 5 patients were assessed as LI-RADS treatment response (LR-TR) viable (TAA in our study) and confirmed as hepatocellular carcinoma by pathological results, 3 patients were assessed as LR-TR nonviable (TNA in our study), no cancerous tissue was seen in 2 patients' pathological results, and in 1 patient's pathological results, the tumor was mostly necrotic with only a small amount of cancerous tissue was seen in the periphery.
In published articles, Kim et al.[1] evaluated viable tumor with a specificity of 98% for both reviewers based on Gd-EOB-MRI images. Jae Seok Bae et al.[2] published an article confirmed that the specificity of hepatobiliary agent-enhanced MRI using the LI-RADS treatment response algorithm (TRA) to assess viable was 93.9%. Yoon et al.[3] showed that the specificity of evaluation as LR-TR viable after transarterial radioembolization treatment was 93.3-100%, and of the lesions evaluated as nonviable with LR-TR, 80.0% (12 of 15) were completely necrotic histopathologically. In Mohammad Chaudhry's study[4], 15 lesions were evaluated as LR-TR viable by LIRADS TRA, 11 of which showed viable tumor on histopathology; 32 lesions were classified as LR-TR nonviable, 26 of which were completely necrotic on histopathology.
Our data and the above literature showed that when LIRADS treatment response is evaluated as viable, it is relatively consistent with the pathological results. When LIRADS treatment response is evaluated as nonviable, Most MR evaluation results correspond to pathology, but a few are inconsistent with pathology. To solve this problem, we finally select TNA areas that are evaluated as nonviable both by DSA and LI-RADS treatment response algorithm. We mentioned this in 2.4 Reference Standard.
We revised the second point (line 288-292) of the limitations in the discussion, hoping that they will not cause ambiguity.
Second, patients in our study did not have pathological results after TACE, so our gold standard is mainly based on TACE retreatment results and LI-RADS standards, which can guarantee the accuracy of the lesion analysis. In the follow-up study, we will focus on collecting patients who have undergone surgical resection after TACE, analyze and discuss the pathological results.
- Kim SW, Joo I, Kim HC, Ahn SJ, Kang HJ, Jeon SK, Lee JM. LI-RADS treatment response categorization on gadoxetic acid-enhanced MRI: diagnostic performance compared to mRECIST and added value of ancillary features. Eur Radiol. 2020;30:2861-70.DOI: 10.1007/s00330-019-06623-9.
- Bae JS, Lee JM, Yoon JH, Kang HJ, Jeon SK, Joo I, Lee KB, Kim H. Evaluation of LI-RADS Version 2018 Treatment Response Algorithm for Hepatocellular Carcinoma in Liver Transplant Candidates: Intraindividual Comparison between CT and Hepatobiliary Agent-enhanced MRI. Radiology. 2021;299:336-45.DOI: 10.1148/radiol.2021203537.
- Yoon J, Lee S, Shin J, Kim SS, Kim GM, Won JY. LI-RADS Version 2018 Treatment Response Algorithm: Diagnostic Performance after Transarterial Radioembolization for Hepatocellular Carcinoma. Korean J Radiol. 2021;22:1279-88.DOI: 10.3348/kjr.2020.1159.
- Chaudhry M, McGinty KA, Mervak B, Lerebours R, Li C, Shropshire E, Ronald J, Commander L, Hertel J, Luo S, et al. The LI-RADS Version 2018 MRI Treatment Response Algorithm: Evaluation of Ablated Hepatocellular Carcinoma. Radiology. 2020;294:320-6.DOI: 10.1148/radiol.2019191581.
Author Response File: 
Author Response.docx
Reviewer 2 Report
In this interesting study, the authors assessed the efficacy of quantitative parameters of intravoxel incoherent motion (IVIM) diffusion-weighted imaging for hepatocellular carcinoma (HCC) diagnosis in 50 HCC patients after transarterial chemoembolization (TACE) and underwent MRI. All patients were scanned with the IVIM-DWI sequence and underwent TACE retreatment within 1 week. Referring to digital subtraction angiography (DSA) and MR-enhanced images, two readers measured the f, D and D* values of the tumor active area (TAA), tumor necrotic area (TNA) and adjacent normal hepatic parenchyma (ANHP). They found that for values of f and D, there were significant differences between any two of the TAA, TNA and ANHP. The values of f and D were the best indicators for identifying the TAA and TNA, with the AUC values of 0.959 and 0.955, respectively. The values of f and D performed well for distinguishing TAA from ANHP, with the AUC values of 0.835 and 0.753, respectively. They concluded that quantitative IVIM-DWI was effective for evaluating tumor viability in HCC patients treated with TACE and may be helpful for non-invasive monitoring of tumor viability.
The study provides original finding with clinical impact. However, there are some issues requiring further data.
-Study population: please provide details on liver function. Were enrolled patients all cirrhotic? Which Child-Pugh class?
-Please add, if availble, alfa-fetoprotein serum levels.
-TACE Treatment: please describe if a selective or superselective TACE procedure was used.
-Discussion: recent literature data demonstrated that efficacy and objective tumor response to TACE, is significantly associated to the transient post-TACE hypertransaminasemia. This is related to the tumor-necrotic effect of TACE, as recenlty demonstrated (TRANS-TACE: Prognostic Role of the Transient Hypertransaminasemia after Conventional Chemoembolization for Hepatocellular Carcinoma. J Pers Med. 2021 Oct 17;11(10):1041. ).
Author Response
Response to Reviewer 2 Comments
In this interesting study, the authors assessed the efficacy of quantitative parameters of intravoxel incoherent motion (IVIM) diffusion-weighted imaging for hepatocellular carcinoma (HCC) diagnosis in 50 HCC patients after transarterial chemoembolization (TACE) and underwent MRI. All patients were scanned with the IVIM-DWI sequence and underwent TACE retreatment within 1 week. Referring to digital subtraction angiography (DSA) and MR-enhanced images, two readers measured the f, D and D* values of the tumor active area (TAA), tumor necrotic area (TNA) and adjacent normal hepatic parenchyma (ANHP). They found that for values of f and D, there were significant differences between any two of the TAA, TNA and ANHP. The values of f and D were the best indicators for identifying the TAA and TNA, with the AUC values of 0.959 and 0.955, respectively. The values of f and D performed well for distinguishing TAA from ANHP, with the AUC values of 0.835 and 0.753, respectively. They concluded that quantitative IVIM-DWI was effective for evaluating tumor viability in HCC patients treated with TACE and may be helpful for non-invasive monitoring of tumor viability.
The study provides original finding with clinical impact. However, there are some issues requiring further data.
Reply: We would like to thank you for the time and effort spent in reviewing the manuscript. And thank you for your comments.
Point 1: Study population: please provide details on liver function. Were enrolled patients all cirrhotic? Which Child-Pugh class?
Response 1: We appreciate your very helpful comments.We supplemented the clinical information of all patients with data on their cirrhosis and Child-Pugh class, all of which were examined within 5 days before TACE treatment, and are added in Table 2. Thank you for enabling us to improve the quality of our manuscript.
Table 2. Patient and lesion characteristics on IVIM
|
Characteristics |
N |
||
|
Sex |
|
||
|
|
Women |
9 |
|
|
|
Men |
41 |
|
|
Age (y) |
54±10 |
||
|
Etiology of HCC |
|
||
|
|
Hepatitis B virus |
40 |
|
|
|
Cryptogenic |
8 |
|
|
|
Hepatitis C virus |
2 |
|
|
|
Alcoholism |
1 |
|
|
Cirrhotic |
|
||
|
|
Yes |
41 |
|
|
|
No |
9 |
|
|
Child-Pugh class* |
|
||
|
|
Child-Pugh A |
41 |
|
|
|
Child-Pugh B |
8 |
|
|
AFP# |
|
||
|
|
Normal |
11 |
|
|
|
Abnormal |
35 |
|
|
Number of HCC lesions per patient |
|
||
|
|
Multiple |
15 |
|
|
|
Single |
35 |
|
|
Number of ROIs |
|
||
|
|
ROIs of the tumor active area |
67 |
|
|
|
ROIs of the tumor necrotic area |
38 |
|
|
|
ROIs of the adjacent normal hepatic parenchyma |
67 |
|
Data are described as the number of patients, lesions and mean ± standard deviation.
*One of the patients lacks the information of Child-Pugh class.
#Four of the patients lacks the information of AFP.
HCC: hepatocellular carcinoma; ROIs: regions of interest. AFP: alpha-fetoprotein
Point 2: Please add, if availble, alfa-fetoprotein serum levels.
Response 2: Thank you very much for your comments.
We supplemented the AFP information of 50 participants, 4 with no AFP information and 46 with AFP results, 11 of whom had AFP values within the normal reference value, and 35 had AFP values above the normal value, all of whom had their AFP examined within 5 days before TACE treatment, and we supplemented the information in Table 2.
Thank the reviewer for valuable suggestions. We are very grateful for the questions raised by the reviewer.
Table 2. Patient and lesion characteristics on IVIM
|
Characteristics |
N |
||
|
Sex |
|
||
|
|
Women |
9 |
|
|
|
Men |
41 |
|
|
Age (y) |
54±10 |
||
|
Etiology of HCC |
|
||
|
|
Hepatitis B virus |
40 |
|
|
|
Cryptogenic |
8 |
|
|
|
Hepatitis C virus |
2 |
|
|
|
Alcoholism |
1 |
|
|
Cirrhotic |
|
||
|
|
Yes |
41 |
|
|
|
No |
9 |
|
|
Child-Pugh class* |
|
||
|
|
Child-Pugh A |
41 |
|
|
|
Child-Pugh B |
8 |
|
|
AFP# |
|
||
|
|
Normal |
11 |
|
|
|
Abnormal |
35 |
|
|
Number of HCC lesions per patient |
|
||
|
|
Multiple |
15 |
|
|
|
Single |
35 |
|
|
Number of ROIs |
|
||
|
|
ROIs of the tumor active area |
67 |
|
|
|
ROIs of the tumor necrotic area |
38 |
|
|
|
ROIs of the adjacent normal hepatic parenchyma |
67 |
|
Data are described as the number of patients, lesions and mean ± standard deviation.
*One of the patients lacks the information of Child-Pugh class.
#Four of the patients lacks the information of AFP.
HCC: hepatocellular carcinoma; ROIs: regions of interest. AFP: alpha-fetoprotein
Point 3: TACE Treatment: please describe if a selective or superselective TACE procedure was used.
Response 3: We appreciate your very helpful comments.We verified the TACE procedure in all patients, most of them were treated with superselective TACE, 2 patients were treated with selective TACE and 48 patients were treated with superselective TACE. we added the information in the results(line 164-165).Thank you for enabling us to improve the quality of our manuscript.
Of the 50 participants, the procedure was elective TACE in 2 patients and super-elective TACE in 48 patients.
Point 4: Discussion: recent literature data demonstrated that efficacy and objective tumor response to TACE, is significantly associated to the transient post-TACE hypertransaminasemia. This is related to the tumor-necrotic effect of TACE, as recenlty demonstrated (TRANS-TACE: Prognostic Role of the Transient Hypertransaminasemia after Conventional Chemoembolization for Hepatocellular Carcinoma. J Pers Med. 2021 Oct 17;11(10):1041. ).
Response 4: Thank you very much for your comments. We have carefully studied and benefited from the reference listed by the reviewer. the necrotic effect after TACE is a research hotspot, which also gives us a good inspiration to remind us to focus on the relationship between patients' clinical manifestations, laboratory tests and imaging manifestations in the next studies, which can help individualize patient treatment as well as assist in guiding clinical decisions.We have added content related to transaminase elevation after TACE in our discussion (line 264-270) and cited this literature, hoping to direct readers and clinicians to pay more attention to the role of transient transaminase elevation after TACE treatment.Thank you for enabling us to improve the quality of our manuscript.
After TACE, the tumor-feeding vessels is embolized, and the tumor becomes ischemic and hypoxic, eventually leading to cellular necrosis, which is manifested in the IVIM sequence by a gradual decrease in f-values and a gradual increase in D-values. The tumor-necrotic effect of TACE may be manifested clinically by fever and pain and an increase in transaminases, and studies have shown that the increase in transaminases after TACE is mainly secondary to tumor necrosis, and that transient post-treatment Transaminase elevation may predict objective response to superselective cTACE in clinical practice[25].
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors gave a lot of explanation on why they did not have the pathology examination and wh the current criteriae should be suffcient. But they did not add the content and the references to the manuscript.
Author Response
Response to Reviewer 1 Comments
Point 1: The authors gave a lot of explanation on why they did not have the pathology examination and why the current criteriae should be suffcient. But they did not add the content and the references to the manuscript.
Response 1: We appreciate your very helpful comments.
In the Materials and Methods ( line 133-136 ) and Discussion ( line 284-300 ), we adjusted and cited literatures to supplement and improve our methods and reference standards. At the same time, we also improved the conclusions( line 333-334 ).
2.4. Reference Standard
LI-RADS treatment response algorithm (TRA) is widely used to evaluate response of HCC to locoregional treatment. However, LI-RADS is not equivalent to pathological findings, and to make the results more accurate, we combined LI-RADS and DSA results. Using MR enhanced images and DSA images as reference standards, two radiologists who had more than 8 years of experience in abdominal diagnosis individually evaluated MR images in accordance with LI-RADS standards[1]( revised version 17) and were unaware of the DSA results. The TAA was defined as LI-RADS-Treatment Response (LR-TR) viable. Two interventionalists with over ten years of experience evaluated the TACE results but were unaware of the MR results. When observing tumor staining, TAAs were considered to exist. Any disagreements were resolved by consulting a senior radiologist. When the area was evaluated as LI-RADS-Treatment Response (LR-TR) viable in LI-RADS TRA and observed staining in DSA, it was considered as TAA; TNA areas was selected when it was evaluated as nonviable both by DSA and LI-RADS TRA.
4.Discussion
Fifty participants in this study had TACE procedure after MR examination without surgical resection and therefore they had no pathological findings. LI-RADS TRA is a widely used method for assessing locoregional treatment response in clinic. Kim et al.[2] ( revised version 25) evaluated viable tumor with a specificity of 98% for both reviewers based on Gd-EOB-MRI images. Jae Seok Bae et al.[3] ( revised version 26)confirmed that the specificity of hepatobiliary agent-enhanced MRI using the LI-RADS treatment response algorithm (TRA) to assess viable was 93.9%. Yoon et al.[4] ( revised version 27) showed that the specificity of evaluation as LR-TR viable after transarterial radioembolization treatment was 93.3-100%, and of the lesions evaluated as nonviable with LR-TR, 80.0% (12 of 15) were completely necrotic histopathologically. In Mohammad Chaudhry's study[5] ( revised version 28), 15 lesions were evaluated as LR-TR viable by LIRADS TRA, 11 of which showed viable tumor on histopathology; 32 lesions were classified as LR-TR nonviable, 26 of which were completely necrotic on histopathology. Those results showed that when LI-RADS treatment response is evaluated as viable, it is relatively consistent with the pathological results. When LI-RADS treatment response is evaluated as nonviable, Most MR evaluation results correspond to pathology, but a few are inconsistent with pathology. We finally combined LI-RADS and DSA results to make the results more accurate.
5.Conclusions
In conclusion, IVIM especially parameter f plays an active role in the quantitative evaluation of the treatment response of hepatocellular carcinoma after TACE and may be an alternative non-invasive method to monitor the tumor viability in patients with HCC after TACE.
Once again, thank you very much for your comments and suggestions.
- Chernyak V, Fowler KJ, Kamaya A, Kielar AZ, Elsayes KM, Bashir MR, Kono Y, Do RK, Mitchell DG, Singal AG, et al. Liver Imaging Reporting and Data System (LI-RADS) Version 2018: Imaging of Hepatocellular Carcinoma in At-Risk Patients. Radiology. 2018;289:816-30.DOI: 10.1148/radiol.2018181494.
- Kim SW, Joo I, Kim HC, Ahn SJ, Kang HJ, Jeon SK, Lee JM. LI-RADS treatment response categorization on gadoxetic acid-enhanced MRI: diagnostic performance compared to mRECIST and added value of ancillary features. Eur Radiol. 2020;30:2861-70.DOI: 10.1007/s00330-019-06623-9.
- Bae JS, Lee JM, Yoon JH, Kang HJ, Jeon SK, Joo I, Lee KB, Kim H. Evaluation of LI-RADS Version 2018 Treatment Response Algorithm for Hepatocellular Carcinoma in Liver Transplant Candidates: Intraindividual Comparison between CT and Hepatobiliary Agent-enhanced MRI. Radiology. 2021;299:336-45.DOI: 10.1148/radiol.2021203537.
- Yoon J, Lee S, Shin J, Kim SS, Kim GM, Won JY. LI-RADS Version 2018 Treatment Response Algorithm: Diagnostic Performance after Transarterial Radioembolization for Hepatocellular Carcinoma. Korean J Radiol. 2021;22:1279-88.DOI: 10.3348/kjr.2020.1159.
- Chaudhry M, McGinty KA, Mervak B, Lerebours R, Li C, Shropshire E, Ronald J, Commander L, Hertel J, Luo S, et al. The LI-RADS Version 2018 MRI Treatment Response Algorithm: Evaluation of Ablated Hepatocellular Carcinoma. Radiology. 2020;294:320-6.DOI: 10.1148/radiol.2019191581.
Author Response File: Author Response.docx
Reviewer 2 Report
The revised manuscript provides the suggested changes and the manuscript can be now accepted.
Author Response
Response to Reviewer 2 Comments
Point 1: The revised manuscript provides the suggested changes and the manuscript can be now accepted.
Response 1: We would like to thank you for the time and effort spent in reviewing the manuscript. We appreciate your positive comments.
Author Response File: Author Response.docx
