Radiological Outcome of Subcentimeter Arterially Enhancing Nodules Detected during Surveillance for Hepatocellular Carcinoma- A Cohort Study
Correspondence Address :
Prabhat Prakash Gupta,
123, Priya Enclave, Near Karkarduma Court, Delhi, Delhi, India.
Introduction: Hepatocellular Carcinoma (HCC) is the sixth most diagnosed cancer and the fourth leading cause of cancer-related death globally. The male: female ratio for HCC in India is 4:1 with age of presentation ranging from 40 to 70 years. There is limited clinical information on the course of Subcentimeter-Sized Nodules (SCSNs) detected during surveillance for HCC.
Aim: To evaluate the serial outcome of subcentimeter arterially enhancing nodules evolving into HCC and henceforth, to identify specific radiological features which can prognosticate and, if possible, predict which SCSNs will turn into HCC.
Materials and Methods: A prospective cohort study was conducted in a tertiary care center in Delhi NCR, in the Department of Radiodiagnosis between 1st May 2018 and 30th April 2019. Total of 72 lesions in 59 patients were evaluated during the study period of one year, which included images spanning over a mean duration of three years (range 2-6.5 years). Dynamic contrast enhanced imaging was done as per Liver Imaging Reporting and Data System version 2018 (LI-RADS version 2018) using either Magnetic Resonance Imaging (MRI) or Computed Tomography (CT). The gold standard for HCC diagnosis was LR 5 lesion. Size cut-off, rate of growth, enhancement features were studied and calculated. Student’s t-test was used for comparison of quantitative outcome parameters.
Results: A total of 59 patients were analysed with mean age 53±12 years, of which 85% were males. The cumulative HCC development rate was 47.5%. A 60.9% of the SCSNs which turned into HCCs showed an increase in size, 31.6% of the non HCC-SCSN lesions also showed an increase in size. Upon baseline comparison, the growth difference was more in the HCC group (8.2±12.24 mm) than in the non HCC group (3.37±7.39 mm). The optimal cut-off points after which the likelihood of an arterially enhancing lesion turning into HCC increased significantly was 8.5 mm on CT and 10.5 mm on MRI.
Conclusion: Around 47.5% of arterially enhancing SCSN converted into HCC; this percentage is much higher than quoted literature. A six monthly follow-up may be considered as 52.2 percent of lesions turned into HCC in a span of 1 year. The optimal cut-off points after which the likelihood of an arterially enhancing lesion turning into HCC increases can be taken as 8.55 mm on CT and 10.5 mm on MRI. Among the lesions showing washout, 80.8 % on CT and 91.3% nodules on MRI changed into HCC.
Computed tomography, Diffusion restriction, Hepatobiliary contrast, Magnetic resonance imaging, Washout
The HCC is the sixth most diagnosed malignancy and the fourth leading cause of cancer-related death globally (1). In India, the age adjusted incidence rate of HCC ranges from 0.7 to 7.5 and 0.2 to 2.2 per 100,000 of population per year for men and women respectively. The male: female ratio for HCC in India is 4:1 with age of presentation ranging from 40 to 70 years (2).
The most reliable diagnostic examinations for diagnosis of HCC are multiphasic CT and MRI, which includes late arterial, portal venous, hepatic venous, and delayed/equilibrium phase imaging at about 3-5 minutes after contrast injection(3),(4). The presence of arterial enhancement followed by washout has a sensitivity of 90%, specificity of 95%, and a positive predictive value of approximately 100% among the group having high chances of developing HCC, e.g., cirrhotics (5). Washout is defined by the appearance of a hypoattenuating area in the portovenous or delayed phase on a triphasic CT liver scan that was hyperdense compared with the rest of the liver in the arterial phase (6).
Hepatobiliary contrast agents are extracellular gadolinium chelates that are now routinely employed in MR liver scanning. Apart from renal excretion, these agents also have a component excreted by the liver depending upon the functional hepatocytes in the parenchyma. A significant benefit of these agents is that they can detect early HCC that shows relative hypoenhancement on the Hepatobiliary Phase (HBP) when there is no appreciable arterial enhancement or venous washout. This can increase the sensitivity and accuracy for HCC diagnosis (7),(8). HBP hypointensity favours a premalignant or a malignant lesion rather than a low-grade dysplastic or a cirrhotic nodule in studies with hepatitis B and C patients (9),(10).
Diffusion-Weighted Imaging (DWI) can enhance the diagnostic performance of MRI for small HCCs by demonstrating higher cellularity of HCCs. Other ancillary imaging features favouring HCC diagnosis include the presence of intralesional fat, mild to modest hyperintensity on T2-weighted images, and morphologic findings such as intratumoural haemorrhage, fatty transformation, and nodule-in-nodule configuration (11).
Accordingly, the European Association for the Study of the Liver and the American Association for the Study of Liver Diseases advocate performing screening for HCC in patients at risk. There are imaging criteria for HCC diagnosis if lesions are 1 cm or larger, but a wait and watch policy is recommended for nodules smaller than 1 cm owing to a high false positive rate. However, there is a high possibility that minute hepatic nodules detected during surveillance may become malignant over time (12),(13),(14). Nevertheless, there is limited information on the clinical course of SCSNs detected during surveillance.
Arterially enhancing lesions are mostly benign and include primary liver tumours like focal nodular hyperplasia, adenoma, and small hemangiomas that rapidly fill with contrast. These have to be differentiated from the most common hypervascular malignant liver tumour, which is HCC; and metastasis from hypervascular neoplasms like melanoma, renal cell carcinoma, breast, sarcoma and neuroendocrine tumours (carcinoid, pheochromocytoma, islet cell tumours).
Meanwhile, according to the LI-RADS classification, which endorses a conservative approach, even though a definitive diagnosis of HCC cannot be established, “probable” HCC can be given for subcentimeter nodules showing one finding among “threshold growth,” “washout,” or “capsule,” (15). Korean Liver Cancer Study Group-National Cancer Center (KLCSG-NCC) practice guidelines suggests that HCCs can be diagnosed by a combination of the typical features of HCCs in ?2 imaging modalities and increased serum ?-fetoprotein levels with an increasing trend over time for liver nodules <1 cm in patients with suppressed hepatitis activity (16).
The diagnosis of subcentimeter HCC can be clinically helpful by providing two different options, i.e., immediate treatment and intense follow-up. The key to achieving long-term survival is promptly implementing effective treatment strategies, including locoregional ablative therapy, hepatic resection, and liver transplantation (17),(18).
Therefore, the current evidence is not concrete in discretely triaging subcentimeter hypervascular nodules in the cirrhotic liver. Also, there is no data from the Indian subcontinent. Furthermore, the rate of growth of SCSN and imaging predictive markers of hypervascular SCSN, which turn into HCC, is also not clear.
This study aimed to evaluate the incidence of arterially enhancing subcentimeter nodules evolving into HCC. To identify any specific size cut-off, rate of growth, or specific radiologic features which can prognosticate and, if possible, predict which subcentimeter arterially enhancing nodule will turn into HCC.
A prospective observational cohort study was conducted in all cirrhotic patients who presented to the Department of Radiodiagnosis at Medanta The Medicity, Gurgaon Haryana, India, for dynamic CT or MRI during the study period from (1st May 2018 to 30th April 2019) The study was undertaken only after the approval from Institutional Review Board.
Inclusion criteria: Patients with liver cirrhosis of any aetiology or chronic liver disease, including chronic hepatitis B and C viral infection, without a prior history of HCC in whom a SCSN was identified during HCC surveillance with either CT or MRI were included.
Exclusion criteria: Patients with lesions size more than 1 cm or known HCC on the first scan were excluded from the analysis. Also, patients who had only a single imaging study were eliminated from the research group.
Sample size calculation: Cochran formula was used for sample size calculation which yielded a minimum number of 51 lesions (19). The final study was done on 72 lesions in 59 patients.
After the inclusion in the study, the patients were further followed for at least one year. The retrospective previous cross-sectional imaging, if available, was also included in the study.
LI-RADS is a system to categorise liver lesions used in patients with liver cirrhosis and chronic hepatitis B without cirrhosis, as these patients are at an increased risk of HCC. LI-RADS staging reflects the probability of HCC and is based on the typical CT and MRI findings in HCC. LI-RADS is currently not to be used in patients <18 years or patients with cirrhosis due to other causes like congenital hepatic fibrosis or vascular disorders. This is because these patients have lower chances of developing HCC (20).
The multiphasic imaging was done in-house CT and MRI scanners:
1. Siemens Somatom Definition Flash- 128 dual source MDCT scanner (Siemens Healthineers, Erlangen, Germany).
2. Siemens Magnetom Verio-3 Tesla (Siemens Healthineers, Erlangen, Germany).
Scans were independently and retrospectively reviewed on a Picture Archiving and Communication System workstation by two Radiologists with more than 10 years’ experience in hepatobiliary imaging, who evaluated the lesions at different time points for above criteria on dynamic CT and MRI. A definitive interobserver disagreement between the assessors, which occurred in 7 cases, required consensus of the radiology team.
The gold standard for HCC diagnosis was LR 5 lesion. An LR 5 lesion is more than 20 mm in size, showing arterial phase hyperenhancement and washout on portovenous phases/enhancing capsule/threshold growth.
The findings were collected in a tabulated form. The following outcomes were assessed: percentage of lesions turning into HCC, size specific cut-off, rate of conversion, and imaging criteria suspicious for HCC.
The analysis involved profiling of patients on different demographic, clinical and radiological findings. Quantitative data sets were presented in terms of means and standard deviation. Qualitative/categorical data were presented as absolute numbers and proportions. Crosstables were constructed and Chi-square test was used for testing of association. Student’s t-test was used for comparing quantitative parameters. The Area Under the Curve (AUC) was measured using ROC curve analysis. The cut of points has been calculated as a minimum of (1-Sensitivity)2 + (1 - Specificity)2. The cut-off values along with corresponding sensitivity & specificity was also calculated. The p-value <0.05 was considered statistically significant. Statistical Package for Social Sciences (SPSS) software version 20.0 was used for statistical analysis.
A total of 59 patients were analysed, of which 85% were males, and 15% were females between ages of 22 to 81 (Mean± standard deviation: 53±12 years). Many of the patients were scanned using both modalities at different time points. Out of 59 patients, 96.6% (n=57) underwent CT examination and 71.2% (n=42) were scanned using MRI. Out of 59 patients, 28 (47.5%) developed HCC during the study period.
Of those who developed HCC, 74.1 % had an increase in size on CT. With MRI as an imaging modality, 58.3% of the HCC lesions showed an increase in size while only 33.3% of the non HCC lesions increased in size (Table/Fig 1).
The nodules’ mean size, when labeled as HCC, was 13.6±15 mm on CT and 18.1±11.8 mm on MRI.
Upon baseline comparison, the growth difference was more in the HCC group (8.2±12.24 mm) than in the non HCC group (3.37±7.39 mm). None of the lesions turning into HCC showed a decrease in size.
There was a substantial increase in size in the nodules that later evolved into HCC compared with nodules that did not. Using CT as an investigation modality, the mean size of an arterially enhancing SCSN at first scanning, which later turned into HCC, was 5.0±3.5 mm, while which did not turn into HCC was 5.4±3.4 mm. Upon continuous follow-up, the lesions that ultimately transformed into HCC had a difference in growth size between first and last CT was of 8.4±14.0 mm vis a vis non HCC group, which had a growth of 1.4±3.6 mm with a p-value of 0.011, which was statistically significant.
The optimal cut-off points after which the likelihood of an arterially enhancing lesion turning into HCC increase was 8.5 mm in CT with a sensitivity of 72.7 % and a specificity of 66.7% (AUC 0.741); and 10.5 mm in MRI with a sensitivity of 77.3 % and a specificity of 66.7% (AUC 0.749) (Table/Fig 2).
Washout was demonstrated in 40.4% of the lesions on CT scan and 50% on MRI scan. Of these, almost 80.8 % of nodules on CT and 87.5% on MRI changed into HCC (p-value <0.001), which is statistically significant (Table/Fig 3).
In patients having nodules that later changed into HCC and were scanned using MRI, 45.8% showed diffusion restriction (p-value 0.002), and 83.3% demonstrated non uptake of hepatobiliary contrast (p-value <0.001), which was statistically significant (Table/Fig 4).
For arterially enhancing SCSN, the mean HCC free interval was 3.3 years and the median HCC free interval was 2.1 years. At a time interval of 1 year, 47.8% of lesions did not change into HCC and at an interval of 4 years, 45.2% of lesions did not convert into malignancy. (The retrospective previous cross-sectional imaging, if available, was also considered in the study. Some of the included patients had previous imaging reaching upto,4-6 years, so their 4 years’ worth of investigation and imaging was included in the study).
In our study, the incidence of development of HCC was 47.5 % from arterially enhancing subcentimeter nodules, with upto 52.2% of lesions changing into HCC within the first year. This percentage is relatively higher than the reported literature; a plausible explanation for this could be the inclusion of cirrhotics or patients of chronic liver disease due to any aetiology rather than patients only of hepatitis B or C virus infection. In the study by Min YW et al., the yearly incidence was 4.9% for the development of HCC from nodules in patients with chronic hepatitis B or C infection (12). Other researchers have also reported a similar incidence of 2-8% for subcentimeter nodules changing into HCC within a year, in patients with chronic hepatitis B or C infection (13),(21). Park MJ et al., reported that small (5-10 mm) arterially enhancing nodules on CT in surveillance for HCC have a 29.5% probability of developing into HCC over a mean 35.7 month of follow-up (22). According to a study by Forner A et al., 15.4% of subcentimeter lesions turned into HCC over a median follow-up period ranging from 23-30 months (23). Jeong YY et al., reported that 13% of arterial enhancing nodules less than 2 cm were HCC in patients with cirrhosis (24). Holland AE et al., who evaluated 45 arterial phase enhancing nodules smaller than 2 cm in cirrhotic livers, found that 35% of patients with cirrhosis had these enhancing nodules and that only 7% of lesions were neoplastic (25). There is wide variation in the current cited literature regarding subcentimeter nodules’ incidence of evolving to malignancy. Still, it is coherent to consider that arterially enhancing lesions show a higher propensity to convert into HCC, as demonstrated in the present study (Table/Fig 5).
The optimal time for follow-up of the patient should be every six months as upto 52.2% of lesions might convert into HCC within one year. Also, lesions can convert into HCC even after remaining stable for 3-4 years. Therefore, hypervascular lesions should be followed-up for at least 6 years (12). Min YW et al., recommended follow-up of SCSNs every few months to detect growth suggestive of malignant transformation (12).
Song KD et al., reported that the initial size of hypervascular SCSN was a significant predictor of progression to overt HCC, with a cut-off value of 5.5 mm (26). As per his findings, most SCSNs (89.9 %) progressed to overt HCCs within 12 months, and all SCSNs larger than 5.5 mm, progressed to overt HCCs within 12 months. Our statistical analysis revealed that the optimal cut-off points after which the likelihood of an arterially enhancing lesion turning into HCC increased was 8.5 mm in CT and 10.5 mm in MRI. So, it concurred that once the lesion nears these cut-off points, they need to be more aggressively followed-up. In addition, in this study, the rate of change of size was a better predictor than baseline size for predicting LR 5 HCC on CT.
In current study, none of the lesions that changed into HCC decreased in size. Of the nodules that changed into HCC over time, only 23.1% on CT and 25% on MRI showed rim enhancement (Table/Fig 6). This is partially explained because when the tumours become 1.5 to 2 cm in diameter, they tend to dedifferentiate, becoming moderately differentiated and growing in an expansile fashion with the formation of a fibrous capsule (27). A surrounding tumour capsule can be detected in lesions of any size, but because of increasing capsule thickness with increasing tumour size, capsules are better seen with bigger HCCs (28). They are typically thin and discontinuous and show progressive delayed enhancement. Capsular enhancement may represent contrast material retained in the fibrous and vascular elements of the tumour margin (29). Capsular enhancement is seldom seen in tumours less than 1 cm in diameter (30). Ishigami K et al., noted that HCCs with histologic fibrous capsules were larger than 1.2 cm in size and that approximately one-half of well-differentiated HCCs had no enhancing capsule (31).
DWI may improve the diagnostic performance of MRI for malignant SCSNs by demonstrating higher cellularity of HCC (11). In this study, 45.8% of small HCCs had restricted diffusion (Table/Fig 6). Albin N et al., elucidated that on DWI, 79% of dysplastic nodules are iso- or hypointense, whereas 97% of HCCs are hyperintense (32).
HBP hypointensity is a strong indicator of premalignancy or malignancy, and its presence favours early HCC over a cirrhotic nodule (10). Out of all who developed HCC, 83.3% of patients showed HBP hypointensity, which was statistically significant (p-value <0.001).
The present study had some limitations. There was a lack of histologic confirmation for the lesions, which were defined on the basis of radiologic images. However, it is unlikely that HCCs were mistakenly categorised, as previous imaging was also considered. Additionaly, pathological validation of these lesions would not be practical in routine clinical settings. Lastly, our assessments regarding the number, nodule pattern, location, and size of SCSNs had a component of subjectivity due to the small nodule sizes and sometimes ill-defined margins. To overcome this limitation, all radiologic images were reviewed by two experienced radiologists. Future large multi-center prospective studies are needed to refine our research findings and confirm our proposed cut-off values.
Thus, around 47.1% of subcentimeter arterially enhancing nodules have the potential to turn into HCC. The size specific cut-off for CT was 8.5 mm, and for MRI was 10.5 mm with a sensitivity of 72.7% and 77.3%. The optimal time for following-up the patient should be every six months as upto 52.2 % of lesions might turn into HCC in less than one year. None of the lesions that changed into HCC decreased in size on imaging. Diffusion restriction and HBP hypointensity are strong predictors of premalignancy or malignancy, and their presence favours early HCC over cirrhotic nodule. This study can be used to power other large multicentric trials with larger sample sizes to further develop recommendations for subcentimeter arterially enhancing nodules in HCC surveillance.
Date of Submission: May 24, 2021
Date of Peer Review: Aug 15, 2021
Date of Acceptance: Feb 15, 2022
Date of Publishing: Jun 01, 2022
• Financial or Other Competing Interests: None
• Was Ethics Committee Approval obtained for this study? Yes
• Was informed consent obtained from the subjects involved in the study? Yes
• For any images presented appropriate consent has been obtained from the subjects. Yes
PLAGIARISM CHECKING METHODS:
• Plagiarism X-checker: May 25, 2021
• Manual Googling: Jan 15, 2022
• iThenticate Software: Feb 11, 2022 (22%)
ETYMOLOGY: Author Origin
- Emerging Sources Citation Index (Web of Science, thomsonreuters)
- Index Copernicus ICV 2017: 134.54
- Academic Search Complete Database
- Directory of Open Access Journals (DOAJ)
- Google Scholar
- HINARI Access to Research in Health Programme
- Indian Science Abstracts (ISA)
- Journal seek Database
- Popline (reproductive health literature)