The number of dysplastic nodules detected clinically has increased since patients with hepatitis virus-associated cirrhosis, who are at increased risk for hepatocellular carcinoma (HCC), began to undergo regular cancer surveillance. Although it is potentially important to determine which type(s) of nodule may be prone to progress to HCC, outcomes of dysplastic nodules have not been fully investigated. This prompted us to examine the outcomes of dysplastic nodules in cirrhotic patients clinicopathologically. We studied 33 dysplastic nodules of <20 mm in maximum diameter, diagnosed by fine needle aspiration biopsy under ultrasonography (US). These nodules were clinically followed, mainly by US examination, for up to 70 months. When the nodules enlarged or exhibited changes on US, they were histologically reexamined by second biopsy. Surprisingly, 15 of the 33 nodules (45.5%) disappeared, 14 nodules (42.4%) remained unchanged, and only 4 nodules (12.1%)progressed to HCC. The latter 4 nodules were all hyperechoic on US and were composed of clear cells with fatty change or small cells with increased nuclear density, and in all 4 patients serum was positive for hepatitis C virus antibody. Univariate analyses revealed that, although not significant, the hyperechoic nodules or nodules with small cell change showed a higher HCC progression rate in comparison with the hypoechoic nodules or the nodules without small cell change. In summary, most of the dysplastic nodules we followed disappeared or remained unchanged, but some progressed to HCC. Hyperechoic nodules in patients with hepatitis C virus-associated cirrhosis, which show small cell change with increased nuclear density, may be prone to progress to HCC.

With the recent advances in imaging modalities in hepatology and regular follow-up examinations of patients with HCV2-associated or HBV-associated chronic liver disease, many sizable nodular lesions have been detected by US (1, 2). It has been reported that 5–7% of patients with HCV-associated or HBV-associated liver cirrhosis progress to HCC per year in Japan (3, 4). When nodules do not exhibit typical features on imaging modalities (5, 6), a biopsy of the nodules is performed for diagnostic purposes. The nodular lesions detected by imaging modalities are recognized as HCC, macroregenerative nodules, dysplastic nodules, or other lesions in biopsy specimens (7, 8). From pathological aspects, formation of a regenerative nodule in cirrhotic liver might be the first step of hepatocarcinogenesis, with subsequent development into an overt HCC nodule through adenomatous hyperplasia,atypical adenomatous hyperplasia, and early HCC in a multistep fashion;alternatively, HCC may arise de novo(9, 10). Not all dysplastic nodules or macroregenerative nodules progress to overt HCC, although some do. However, the clinical outcomes of dysplastic nodules have not been determined clearly.

Here we studied 33 sizable nodules that had been diagnosed as dysplastic nodules in our university hospital over an ∼10-year period and examined the outcomes of the nodules clinicopathologically. We found that most of the dysplastic nodules disappeared or remained unchanged, but that dysplastic nodules with several particular features might be prone to progress to HCC.

Materials.

We studied 33 dysplastic nodules obtained by fine needle (21-gauge)biopsy from 33 cirrhotic patients who had been examined mainly by US in addition to determination of AFP or protein induced by vitamin K absence or antagonist-II (PIVKA-II) in serum every 3 or 4 months as individuals at high risk for development of HCC between 1987 and 1996. The characteristics of these patients are summarized in Table 1. The dysplastic nodules observed on US were classified into three patterns (US pattern): hyperechoic pattern in 16 patients, hypoechoic pattern in 16 patients, and combined hyper-hypoechoic pattern in 1 patient. In some cases, other imaging modalities such as CT, MRI, or angiography were performed. The biopsy procedure consisted of two punctures of one lesion. The extranodular liver tissue was also obtained at the same time as a control. Informed consent was obtained from each patient.

Histological Evaluation.

All nodules investigated were <20 mm in diameter and were histologically diagnosed as low- or high-grade dysplastic nodules according to the recommendation of the International Working Party (7) by two pathologists at our university hospital. Biopsied nodules were diagnosed as dysplastic nodules when they showed atypical features not diagnostic of carcinoma and were further classified into low- or high-grade dysplastic nodules. Low-grade dysplastic nodules were composed of hepatocytes that were minimally abnormal. The nuclear:cytoplasmic ratio was normal or slightly increased, nuclear atypia was mild, cytoplasm was clear or contained fat, and the liver cell plates were one to two cells wide but focally appeared wider. On the other hand, high-grade dysplastic nodules were composed of hepatocytes that were moderately abnormal. Cell density increased to >1.3 times but less than twice normal, nuclei had irregular contours or exhibited hyperchromasia, microacinar formation appeared in places, and cytoplasmic basophilia or cytoplasmic clear cell change was found.

Clinical Follow-Up.

We followed these patients as very high-risk individuals for HCC every 3 or 4 months, mainly by US. When the dysplastic nodules changed in size or US pattern, we performed a biopsy again and investigated the outcome of the nodules. Observation periods ranged from 4 to 70 months.

Statistical Analysis.

The rates of HCC progression in dysplastic nodules were obtained by the Kaplan-Meier method, and differences in curves were determined by the log-rank test. P < 0.05 was considered significant.

Clinical Outcomes of Dysplastic Nodules.

The median age of the 33 patients was 58 years (range, 36–73),and the male:female ratio was 23:10. The causative agent of chronic liver disease associated with dysplastic nodules was HCV in 26 patients(79%), HBV in 6 patients (18%), and non-B, non-C agent(s) in one patient (3%). All 33 patients were classified as Child-Pugh stage A. The maximum size of the dysplastic nodules on US ranged from 6 to 20 mm. These data are summarized in Table 1. The median follow-up period was 25 months (range, 4–70), and no patients died during the study. Fifteen of the 33 dysplastic nodules (46%) disappeared between 4 and 70 months of follow-up. Disappearance of the nodules was strictly confirmed by at least two consecutive US examinations. Seven of these 15 nodules were hyperechoic, 7 were hypoechoic, and 1 had a combined hyper-hypoechoic pattern on US. Fourteen of the 33 nodules(42%) remained unchanged in size over 8–46 months. Five of these 14 nodules were hyperechoic, whereas the remaining 9 were hypoechoic on US. Four of the 33 nodules (12%) progressed to HCC over 16–21 months. All four of these nodules showed hyperechoic pattern on US. They were 11, 13, 16, and 20 mm in greatest diameter, and the male:female ratio was 1:3. The median age of the four patients was 63 years, and the causative agent of the associated chronic liver disease was HCV in all four. The values of AFP did not exceed 200 ng/ml nor did those of PIVKA-II exceed 0.06 IU/ml. In 8 patients (nos. 3, 6, 7, 13, 15, 23,24, and 27), HCC suddenly appeared in the areas distant from the identified dysplastic nodule during the follow-up periods. These HCCs were confirmed by biopsy.

Histological Characteristics.

The hepatocytes in the dysplastic nodules were arranged in a trabecular pattern in all patients. Nuclear crowding less than twice that in surrounding control areas (58%), an increase in nuclear:cytoplasmic ratio (45%), and clear hepatocytes (70%; clear cells, 40%;basophilic + clear cells, 15%; and acidophilic + clear cells, 15%)were found frequently. Fatty change (39%), small cell change (52%),and large cell change (27%) were often found. Eight nodules (nos. 1,2, 3, 4, 14, 16, 17, and 33; Table 1) were classified as low-grade dysplastic nodules because nuclear crowding, increased nuclear:cytoplasmic ratio, or enlarged nuclei were not particularly conspicuous, and the remaining 25 were high-grade dysplastic nodules. Four patients (nos. 30–33) progressed to HCC. Patient no. 30 developed HCC in the 17th month after the first biopsy. In the first biopsy specimen from this patient, partial nuclear crowding, clear cell change, and anisocytosis with hyperchromatic nuclei were found, and the lesion was diagnosed as a high-grade dysplastic nodule (Fig. 1,A). Seventeen months later,this nodule had progressed to HCC, as confirmed by second biopsy (Fig. 1,B). The lesion was classified as well-differentiated HCC and was composed of small acidophilic cancer cells. The nuclear:cytoplasmic ratio was increased, and the nuclear density was more than twice that of the control specimen (neighboring cirrhotic area). In the other three patients, diagnoses of HCC were also histologically confirmed. In the first biopsy specimens for these patients, increased nuclear:cytoplasmic ratio, fatty change, nuclear crowding, small cell change, clear cell change, and enlarged or hyperchromatic nuclei were very mild in the hepatocytes of the nodules(Table 1). In the second biopsy, some mitotic figures, many clear cells, marked fatty change, increase in nuclear density to twice or more that in the control area, small hepatocytes, and increased nuclear:cytoplasmic ratio were found. Therefore, the diagnosis was well-differentiated HCC.

Statistical Analysis.

Univariate analyses of prognostic clinicopathological factors were performed. Although not statistically significant, dysplastic nodules that were hyperechoic on US or showing small cell change had a higher malignant progression rate (Fig. 2). Other factors such as nuclear crowding and large cell change did not influence the outcomes of dysplastic nodules (Table 2).

The dysplastic nodules were detected mainly by US but were not always detected by other imaging modalities including CT, MRI, or angiography. Even on MRI, the nodules did not yield typical images of HCC, nor did they on dynamic CT or angiography. In addition, in all patients the values of AFP were <200 ng/ml and not indicative of malignancy (11). The values of PIVKA-II were within normal limits in all patients. We therefore performed a biopsy of those nodules for diagnostic purposes. Although the biopsied lesions might not always be representative of the nodules, it was surprising that 15 of the 33 nodules (45.5%) disappeared, 14 nodules (42.4%) remained unchanged in size and in US features, and only 4 nodules (12.1%)progressed to HCC during follow-up periods. This is in line with a previous observation of Terasaki et al.(12) in that 25 of 34 nodules became undetectable on imaging modalities during the clinical follow-up periods. Although the fact that only a small percentage of dysplastic nodules progress to HCC may not be well recognized by clinicians currently, an analogous phenomenon in experimental hepatic carcinogenesis in rats was reported by Farber and Cameron (13) >20 years ago.

Regarding the findings on US, 16 of the 33 nodules exhibited a hyperechoic pattern, and 4 of these 16 nodules were proved to have progressed to HCC. On the other hand, 16 nodules with hypoechoic pattern on US did not progress to HCC during the follow-up period. Comparison of the HCC progression rate in nodules with hyperechoic or hypoechoic patterns suggested that the hyperechoic nodules might be at higher risk for HCC progression, but this was not statistically significant. Previous reports from our laboratory (3) and others (14, 15, 16, 17, 18) have demonstrated that hyperechoic pattern on US appeared to reflect predominantly fatty change, clear cell change, and small cell change with increased nuclear crowding. In the present study, atypia of hepatocytes in the dysplastic nodules varied from mild to severe, as in the study by Terada et al.(19). Some nodules composed of hepatocytes with some atypical features disappeared, whereas some nodules remained unchanged. However, those dysplastic nodules that progressed to HCC frequently contained small hepatocytes with nuclear crowding, clear hepatocytes,and hepatocytes with fatty change in our study, as well as the recent study by Terasaki et al.(12), who reported that histological features predicting malignant transformation of nonmalignant hepatocellular nodules were an increase in nuclear density, small cell change, and fatty change. Therefore, considering both the US pattern and histological features of dysplastic nodules, it appears that hyperechoic dysplastic nodules composed of small cell change with increased nuclear density may be prone to progress to HCC. Additional studies are needed to evaluate the clinicopathological significance of these morphological changes in dysplastic nodules and to confirm our hypothesis.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

                
2

The abbreviations used are: HCV, hepatitis C virus; HBV, hepatitis B virus; US, ultrasonography; HCC, hepatocellular carcinoma; CT, computed tomography; MRI, magnetic resonance imaging;AFP, α-fetoprotein.

Fig. 1.

Dysplastic nodule progressed to HCC 17 months after the first biopsy (patient no. 30). A, first biopsy. Small cell change with increased nuclear density, clear cell change, and anisocytosis were found. A high-grade dysplastic nodule is shown. H&E staining, ×150. B, second biopsy 17 months after the first biopsy. Small acidophilic cells with nuclear density twice that in neighboring areas and dilated capillaries were observed. A well-differentiated HCC is shown. H&E staining, ×150.

Fig. 1.

Dysplastic nodule progressed to HCC 17 months after the first biopsy (patient no. 30). A, first biopsy. Small cell change with increased nuclear density, clear cell change, and anisocytosis were found. A high-grade dysplastic nodule is shown. H&E staining, ×150. B, second biopsy 17 months after the first biopsy. Small acidophilic cells with nuclear density twice that in neighboring areas and dilated capillaries were observed. A well-differentiated HCC is shown. H&E staining, ×150.

Close modal
Fig. 2.

Effect of ultrasonography patterns(A) and small cell change (B) on HCC progression rate (log-rank test).

Fig. 2.

Effect of ultrasonography patterns(A) and small cell change (B) on HCC progression rate (log-rank test).

Close modal
Table 1

Clinicopathological findings for dysplastic nodules

No.aClinical dataPathological findingscOutcome of follow-up period
Age/SexVirusTumor sizeUS patternbNuclear crowdingCytoplasmic stainingFatty changeIncreased N:C ratioNucleusSCC/LCC
HyperchromasiaIrregular contour
41 /M HCV 9 mm Hyper − − − − − −/− DA (70 mo) 
60 /M HCV 10 Hyper − − − − −/− DA (5) 
38 /M HBV 10 Hyper − − − − − −/− DA (35) 
68 /F HCV 13 Hyper − − − − −/+ DA (10) 
57 /F HCV 14 Hyper B+ C − +/+ DA (68) 
60 /M HCV 15 Hyper A+ C − − − − ++/− DA (10) 
54 /M HCV 18 Hyper − ++ − − − −/+ DA (38) 
58 /F HCV Hypo − +/− DA (44) 
51 /F HCV 12 Hypo − − − − −/− DA (25) 
10 73 /M HCV 13 Hypo B+ C − +/− DA (8) 
11 49 /M HCV 13 Hypo − B+ C − − − − −/− DA (55) 
12 56 /M HCV 16 Hypo − A+ C − −/+ DA (4) 
13 63 /M HCV 20 Hypo − ++/− DA (14) 
14 52 /M HCV 20 Hypo − − − − − −/+ DA (14) 
15 61 /M HCV 16 Com A+ C +/+ DA (66) 
16 59 /F HCV Hyper − − − − − −/+ NC (41) 
17 58 /M HCV Hyper − − − − − −/− NC (13) 
18 68 /F HCV Hyper B+ C +/− NC (31) 
19 64 /M HCV 11 Hyper − A+ C − − − −/+ NC (28) 
20 39 /M HBV 12 Hyper B+ C +/− NC (21) 
21 49 /M HCV Hypo − − − +/− NC (19) 
22 45 /M HBV Hypo A+ B − +/− NC (26) 
23 60 /M HBV 11 Hypo A+ B − − − − −/− NC (15) 
24 63 /M HCV 12 Hypo − +/− NC (36) 
25 69 /M NBNC 13 Hypo − +/− NC (34) 
26 49 /M HCV 15 Hypo − − − −/− NC (8) 
27 61 /F HCV 16 Hypo − − − − −/+ NC (38) 
28 63 /M HBV 16 Hypo − − − − −/− NC (9) 
29 36 /M HBV 17 Hypo − − − +/− NC (46) 
30 65 /F HCV 16 Hyper − +/− HCC (16) 
31 71 /M HCV 11 Hyper − +/− HCC (21) 
32 61 /F HCV 13 Hyper − +/− HCC (17) 
33 58 /F HCV 20 Hyper − A+ C − − − +/− HCC (34) 
No.aClinical dataPathological findingscOutcome of follow-up period
Age/SexVirusTumor sizeUS patternbNuclear crowdingCytoplasmic stainingFatty changeIncreased N:C ratioNucleusSCC/LCC
HyperchromasiaIrregular contour
41 /M HCV 9 mm Hyper − − − − − −/− DA (70 mo) 
60 /M HCV 10 Hyper − − − − −/− DA (5) 
38 /M HBV 10 Hyper − − − − − −/− DA (35) 
68 /F HCV 13 Hyper − − − − −/+ DA (10) 
57 /F HCV 14 Hyper B+ C − +/+ DA (68) 
60 /M HCV 15 Hyper A+ C − − − − ++/− DA (10) 
54 /M HCV 18 Hyper − ++ − − − −/+ DA (38) 
58 /F HCV Hypo − +/− DA (44) 
51 /F HCV 12 Hypo − − − − −/− DA (25) 
10 73 /M HCV 13 Hypo B+ C − +/− DA (8) 
11 49 /M HCV 13 Hypo − B+ C − − − − −/− DA (55) 
12 56 /M HCV 16 Hypo − A+ C − −/+ DA (4) 
13 63 /M HCV 20 Hypo − ++/− DA (14) 
14 52 /M HCV 20 Hypo − − − − − −/+ DA (14) 
15 61 /M HCV 16 Com A+ C +/+ DA (66) 
16 59 /F HCV Hyper − − − − − −/+ NC (41) 
17 58 /M HCV Hyper − − − − − −/− NC (13) 
18 68 /F HCV Hyper B+ C +/− NC (31) 
19 64 /M HCV 11 Hyper − A+ C − − − −/+ NC (28) 
20 39 /M HBV 12 Hyper B+ C +/− NC (21) 
21 49 /M HCV Hypo − − − +/− NC (19) 
22 45 /M HBV Hypo A+ B − +/− NC (26) 
23 60 /M HBV 11 Hypo A+ B − − − − −/− NC (15) 
24 63 /M HCV 12 Hypo − +/− NC (36) 
25 69 /M NBNC 13 Hypo − +/− NC (34) 
26 49 /M HCV 15 Hypo − − − −/− NC (8) 
27 61 /F HCV 16 Hypo − − − − −/+ NC (38) 
28 63 /M HBV 16 Hypo − − − − −/− NC (9) 
29 36 /M HBV 17 Hypo − − − +/− NC (46) 
30 65 /F HCV 16 Hyper − +/− HCC (16) 
31 71 /M HCV 11 Hyper − +/− HCC (21) 
32 61 /F HCV 13 Hyper − +/− HCC (17) 
33 58 /F HCV 20 Hyper − A+ C − − − +/− HCC (34) 
a

Nos. 1–4, 14, 16, 17, and 33 are low-grade dysplastic nodules; others are high-grade dysplastic nodules.

b

Hyper, hyperechoic; Hypo,hypoechoic; Com, combined hyper- and hypoechoic; NC ratio,nuclear:cytoplasm ratio; SCC, small cell change; LCD, large cell change; C, clear cells; A, acidophilic cells; mo, months; B, basophilic cells; DA, disappearance of nodule; NC, no change in nodule.

c

+, present; ++, many present; −,absent.

Table 2

Result of univariate analysis: prognostic clinicopathological factors of malignant transformation

P
US pattern (hyperechoic vs. hypoechoic) 0.064 
Small cell change 0.099 
Large cell change 0.152 
Cytoplasmic staining 0.165 
Fatty change 0.192 
Irregular contour of nucleus 0.236 
Increased N:C ratio 0.275 
Nuclear crowding 0.470 
Hyperchromasia 0.498 
P
US pattern (hyperechoic vs. hypoechoic) 0.064 
Small cell change 0.099 
Large cell change 0.152 
Cytoplasmic staining 0.165 
Fatty change 0.192 
Irregular contour of nucleus 0.236 
Increased N:C ratio 0.275 
Nuclear crowding 0.470 
Hyperchromasia 0.498 

We thank Dr. Mikimi Sakurai (Professor Emeritus of Osaka City University Medical School) and Dr. Kenichi Wakasa (Associate Professor of Clinical Pathology, Osaka City University Hospital) for their useful suggestions.

1
Okuda K. Hepatocellular carcinoma: recent progress.
Hepatology
,
15
:
948
-963,  
1992
.
2
Bisceglie A. M. D., Carithers R. L., Jr., Gores G. J. Hepatocellular carcinoma.
Hepatology
,
28
:
1161
-1165,  
1998
.
3
Oka H., Kurioka N., Kim K., Kanno T., Kuroki T., Mizoguchi Y., Kobayashi K. Prospective study of early detection of hepatocellular carcinoma in patients with cirrhosis.
Hepatology
,
12
:
680
-687,  
1990
.
4
Tsukuma H., Hiyama T., Tanaka S., Nakao M., Yabuuchi T., Kitamura T., Nakanishi K., Fujimoto I., Inoue A., Yamazaki H., Kawashima T. Risk factors for hepatocellular carcinoma among patients with chronic liver disease.
N. Engl. J. Med.
,
328
:
1797
-1801,  
1993
.
5
Takayasu K., Furukawa H., Wakao F., Muramatsu Y., Abe H., Terauchi T., Winter T. C., III, Sakamoto M., Hirohashi S. CT diagnosis of early hepatocellular carcinoma: sensitivity, findings, and CT-pathologic correlation.
Am. J. Roentgenol.
,
164
:
885
-890,  
1995
.
6
Oliver J. M. , III, and Baron, R.
L. Helical biphasic contrast-enhanced CT of the liver: technique
,
indications,interpretation,andpitfalls.Radiology,20
:
1
-14,  
1996
.
7
International Working Party. Terminology of nodular hepatocellular lesions. Hepatology, 22: 983–993, 1995.
8
Ferrell L. D., Crawford L. M., Dhillon A. P., Scheuer P. J., Nakanuma Y. Proposal for standardized criteria for the diagnosis of benign, borderline, and malignant hepatocellular lesions arising in chronic advanced disease.
Am. J. Surg. Pathol.
,
17
:
1113
-1123,  
1993
.
9
Sakamoto M., Hirohashi S., Shimosato Y. Early stages of multistep hepatocarcinogenesis: adenomatous hyperplasia and early hepatocellular carcinoma.
Hum. Pathol.
,
22
:
172
-178,  
1991
.
10
Kondo F., Ebara M., Sugiura N., Wada K., Kita K., Hirooka N., Nagato Y., Kondo Y., Ohto M., Okuda K. Histological features and clinical course of large regenerative nodules: evaluation of their precancerous potentiality.
Hepatology
,
12
:
592
-598,  
1990
.
11
Ebara M., Ohto M., Shinagawa T., Sugiura N., Kimura K., Matsutani S., Morita M., Saisho H., Tsuchiya Y., Okuda K. Natural history of minute hepatocellular carcinoma smaller than three centimeters complicating cirrhosis.
Gastroenterology
,
90
:
280
-298,  
1986
.
12
Terasaki S., Kaneko S., Kobayashi K., Nonomura A., Nakanuma Y. Histological features predicting malignant transformation of nonmalignant hepatocellular nodules: a prospective study.
Gastroenterology
,
115
:
1216
-1222,  
1998
.
13
Farber E., Cameron R. The sequential analysis of cancer development.
Adv. Cancer Res.
,
31
:
125
-226,  
1980
.
14
Borzio M., Borzio F., Croce A., Sala M., Salmi A., Leandro G., Bruno S., Roncalli M. Ultrasonography-detected macroregenerative nodules in cirrhosis: a prospective study.
Gastroentrology
,
112
:
1617
-1623,  
1997
.
15
Liver Cancer Study Group of Japan. Part 3, histological findings. In: Classification of Primary Liver Cancer, pp. 27–73. Tokyo: Kanehara Co., 1997.
16
Sugihara S., Nakashima O., Kojiro M., Majima Y., Tanaka M., Tanikawa K. The morphologic transition in hepatocellular carcinoma.
Cancer (Phila.)
,
70
:
1488
-1492,  
1992
.
17
Tanaka S., Kitamura T., Imaoka S., Sasaki Y., Taniguchi H., Ishiguro S. Hepatocellular carcinoma: sonographic and histologic correction.
Am. J. Roentgenol.
,
140
:
701
-707,  
1983
.
18
Edamitsu O., Kiyomatsu K., Nakashima O., Sugihara S., Kojiro M., Saitsu H., Nakayama T., Majima Y., Tanikawa K. Pathomorphologic study on hepatocellular carcinoma showing hyperechoic pattern.
Acta Hepatol. Jpn.
,
32
:
618
-624,  
1991
.
19
Terada T., Ueda K., Nakanuma Y. Histopathological and morphometric analysis of atypical adenomatous hyperplasia of human cirrhotic livers.
Virchows Arch. A Pathol. Anat.
,
422
:
381
-388,  
1993
.