Increased Liver Iron Stores in Patients With Hepatocellular Carcinoma Developed on a Noncirrhotic Liver BRUNO TURLIN, 1'3 FRI~DI~RIC JUGUET, 2 ROMAIN MOIRAND, 2'3 DANIELLE LE QUILLEUC, 2 OLIVIER LORI~AL,3 JEAN-PIERRE CAMPION, 4 BERNARD LAUNOIS, 4 MARIE-PAULE RAMI~E, 1 PIERRE BRISSOT, 2'3 AND YVES DEUGNIER, 2'3
Iron w a s s y s t e m a t i c a l l y s t u d i e d in the n o n t u m o r o u s liver o f 24 patients w i t h h e p a t o c e l l u l a r c a r c i n o m a (HCC) d e v e l o p e d o n a n o n c i r r h o t i c liver c o m p a r e d w i t h 4 control groups (cirrhosis w i t h a n d w i t h o u t HCC, liver metastasis, a n d n o r m a l liver) m a t c h e d a c c o r d i n g to age, sex, a n d p r e s e n c e of c h r o n i c alcoholism. A s s e s s m e n t of liver iron w a s m a d e by (1) h i s t o l o g y a c c o r d i n g to iron distribution a n d quantification (total iron score: 0 to 60), a n d (2) b i o c h e m i s t r y (liver iron c o n c e n t r a t i o n - N < 36 pmol/g) w i t h c a l c u l a t i o n o f t h e h e p a t i c iron index (liver iron concentration/age). P a t i e n t s w i t h h e p a t o c e i l u l a r c a r c i n o m a d e v e l o p e d o n a n o n c i r r h o t i c liver p r e s e n t e d w i t h (1) h i s t o l o g i c a l i r o n i n 83%; (2) p a r e n c h y m a l
cases. We hypothesized that this iron excess might b e r e l a t e d to H C C i n so f a r a s h u m a n 1"s a n d e x p e r i m e n t a l 9-1~ d a t a s u g g e s t t h a t i r o n p l a y s a r o l e i n t h e d e v e l o p ment of malignancy. Therefore, we documented systematically liver iron stores in a series of patients o p e r a t e d for H C C / N C L i n o r d e r to t e s t t h e h y p o t h e s i s of an iron excess in such patients. Then, the results of these iron studies were compared with those of four matched control groups (cirrhosis with and without HCC, liver metastasis, and normal liver).
iron
PATIENTS AND METHODS
excess significantly m o r e f r e q u e n t (90%) than in controis; (3) total iron score (15 _+ 12) a n d liver iron c o n c e n tration (81 _+ 96) significantly greater t h a n in controls; a n d (4) hepatic iron i n d e x significantly i n c r e a s e d (1.4 _+ 1.5) w h e n c o m p a r e d w i t h control groups, except for the h e p a t o c e l l u l a r c a r c i n o m a c o m p l i c a t i n g cirrhosis g r o u p (0.9 _+ 1.1). This study (1) s h o w s a mild but unquestionable p a r e n c h y m a l iron e x c e s s in t h e n o n t u m o r o u s liver o f m o s t p a t i e n t s p r e s e n t i n g w i t h h e p a t o c e l l u l a r c a r c i n o m a d e v e l o p e d o n a n o n c i r r h o t i c liver and, at a lesser extent, o n cirrhosis, (2) s h o u l d incite o t h e r s to study t h e p u t a t i v e role o f i r o n in t h e d e v e l o p m e n t o f liver c a n c e r b o t h in patients w i t h cirrhosis a n d t h o s e w i t h o u t it, w h a t e v e r t h e c a u s e o f the u n d e r l y i n g liver disease, a n d (3) add a r g u m e n t to t a k e into a c c o u n t a n d to treat a n y liver iron excess, e v e n w h e n mild. (HEPATOLOGY 1995;22:446-450.) When examining the nontumorous part of the liver of patients with hepatocellular carcinoma developed on a noncirrhotic liver (HCC/NCL), we were impressed by the finding of a significant iron overload in most Abbreviations:HCC,hepatocellularcarcinoma;NCL,noncirrhoticliver;META, metastasis;LIC,liver iron concentration;CIR,cirrhoticliver,HBc,hepatitis B core antigen; ELISA,enzyme-linkedimmunosorbentassay; HCV, hepatitis C virus. From the 'Laboratoire d'Anatomie Pathologique B; 2Cliniquedes Maladies du Foie;3INSERMUnite de RecherchesHepatologiquesU49; 4CliniqueChirurgicale B, Hopital Pontchaillou, Rennes, France. Received November 9, 1994; accepted March 3, 1995. Supported by the Fondation pour la Recherche Medicale and the Faculte de Medecine de Rennes (BT) and by the Association pour la Recherche contre le Cancer and the Soci~teNationale Francaise de Gastro-Ent~rologie(YD). Address reprint requests to: Bruno Turlin, MD, Laboratoire d'Anatomie Pathologique B, Hopital Pontchaillou, 35033 Rennes, France. Copyright © 1995 by the American Association for the Study of Liver Diseases. 0270-9139/95/2202-001153.00/0
S a m p l e Selection The s t u d y series consisted of 24 surgical nontumorous liver s a m p l e s removed from 24 p a t i e n t s p r e s e n t i n g with HCC/ NCL. Selection of the s a m p l e s was m a d e by reviewing the l a s t 143 cases of liver resection performed for HCC a t the D e p a r t m e n t of S u r g e r y between 1983 a n d 1992. Of the 114 of 143 cases in which a large ( > 1 × 1 cm) s a m p l e o f n o n t u m o r ous liver was available, 24 (21%) p r e s e n t e d with a definitively noncirrhotic liver. F o u r m a t c h e d control series were selected from the files of the D e p a r t m e n t of Pathology a n d consisted of large ( > 1 × 1 cm) surgical nontumorous liver s a m p l e s from 88 p a t i e n t s p r e s e n t i n g with e i t h e r HCC developed in a cirrhotic liver (HCC/CIR, n = 20), or cirrhosis noncomplicated by HCC (CIR, n = 24), or noncirrhotic liver with m e t a s t a s i s (META, n = 20) from colon (n = 17), p a n c r e a s (n = 1), lung (n = 1) or stomach (n = 1) cancer, or n o r m a l liver t a k e n d u r i n g l a p a r o t o m y for cholecystectomy or esophageal surgery (n = 24). M a t c h i n g was performed according to age, sex, a n d the presence or absence of chronic alcoholism. Two of these four control series were incomplete because of seven unm a t c h a b l e young women (n = 3) a n d old m e n (n = 4).
Data Recorded Clinical, Biological, and Imaging Data. Age, sex, and alc0hol i n t a k e were sought from p a t i e n t s ' c h a r t s in all cases. Chronic alcoholism was defined as the consumption of more t h a n 80 g of alcohol for more t h a n 10 years. In addition, the following d a t a were recorded w h e n available, in HCC/NCL patients: p a s t t r a n s f u s i o n (more t h a n 7 y e a r s before t h e diagnosis of HCC), tobacco smoking (more t h a n 10 cigarettes per d a y for more t h a n 10 years), s t a y in endemic a r e a for viral hepatitis, ascites, a n t i - h e p a t i t i s B core a n t i g e n (HBc) antibody, a n t i - h e p a t i t i s C virus (HCV) antibody (second generation enzymes-linked i m m u n o s o r b a n t a s s a y [ELISA] test), liver function tests, s e r u m a l p h a fetoprotein (n < 10 ng/mL),
446
HEPATOLOGYVol. 22, No. 2, 1995
TURLIN ET AL
TABLE 1. C l i n i c a l a n d B i o l o g i c a l D a t a f r o m 24 P a t i e n t s W i t h HCC Developed on a Noncirrhotic Liver
Gender (M/F) Age (yr) Chronic alcoholism Anti-HBc antibody Anti-HVC antibody Tobacco smoking Past transfusion Esophageal varices Ascites Portal thrombosis Prothrombin time (%) AST (× ULN) ALT (× ULN) ALP (× ULN) GGT (× ULN) Serum iron (#mol/L) Transferrin saturation (%)
447
RESULTS
Characteristics of Patients
No. of Patients
Mean _+ SD or No. of Patients (+)
Range or Percentage
24 24 24 22 6 22 19 23 24 23 24 24 24 24 24 16 16
21/3 55 _+ 18 15 2 0 16 11 1 3 7 92 _+ 11 1.4 +_ 1.2 1.1 _+ 0.9 1.6 +_ 1.9 3.4 +_ 4.6 16 _+ 9 30 _+ 22
(16-79) (62%) (9%) (0%) (73%) (58%) (4%) (12%) (30%) (60-100) (0.3-5.2) (0.2-4.5) (0.5-9.6) (0.6-21) (3-34) (5-85)
Abbreviations: AST, aspartate transaminase; ALT, alanine transaminase; ALP, alkaline phosphatase; GGT, gamma glutamyl transpeptidase; ULN, upper limit of normal.
esophageal varices on endoscopy, a n d portal thrombosis on abdominal u l t r a s o u n d examination. Liver Iron Concentration Measurement. Liver iron concent r a t i o n (LIC) was d e t e r m i n e d according to Barry a n d Sherlock 14 on samples weighing more t h a n 0.5 mg. 1~ Results were expressed as #mol/g dry weight (n < 36). The hepatic iron index, defined as the ratio of liver iron concentration to age (LIC/age, #mol/g per year) was calculated according to Bassett et al. TM Pathological Data. HCC size was defined as the largest diameter of tumor. I n case of multiple HCC, the diameter of the largest t u m o r was t a k e n into account. All specimens were routinely processed for histology, i.e., fixed in n e u t r a l formalin, embedded in paraffin, cut in 4-#m thick slices a n d stained with hematoxylin-eosin-saffron, Sirius red, a n d Perls' staining. HCC were classified according to Edmondson's grade. 17 Fibrosis was graded as follows: 0 = no fibrosis; 1 = portal fibrosis without septa; 2 = septal fibrosis without bridging; 3 = bridging fibrosis without cirrhosis; a n d 4 = cirrhosis as u s u a l l y defined. TM With respect to liver iron, three p a t t e r n s were defined: (1) no iron visible at ×20 magnification; (2) p a r e n c h y m a l iron overload characterized by a p r e d o m i n a n t l y hepatocytic a n d periportal deposition of iron with a decreasing g r a d i e n t t h r o u g h o u t the lobule; and, (3) m e s e n c h y m a l iron overload characterized by iron deposition m a i n l y w i t h i n sinusoidal cells. The histological quantification of iron was made according to Deugnier et al ~9by scoring iron separately within hepatocytes (Hepatocytic Iron Score, 0 to 36), sinusoidal cells (Sinusoidal Iron Score, 0 to 12), a n d portal tracts or fibrosis (Portal Iron Score, 0 to 12). The Total Iron Score (0 to 60) was defined by the s u m of these three scores. Statistical Methods. Results were expressed as m e a n _+ SD (range) a n d as a n u m b e r (%). X2, Fisher exact test, a n d n o n p a r a m e t r i c tests of M a n n - W h i t n e y , Kruskall-Wallis, a n d Wilcoxon were used as appropriate. A P value < .05 was considered significant.
The HCC/NCL patients consisted of 21 men (87%) ranging in age from 31 to 79 years (61 _+ 12) and 3 young women aged 15, 19, and 21 years, respectively. These patients did not present with significant hepatic failure (Table 1). Only 1 of 23 exhibited esophageal varices on endoscopic examination. Portal thrombosis was found in 7 cases, 3 of which presented with ascites. Most patients were chronic alcoholics (62%) and were or had been smokers (73%). A few patients (9%) were positive for the anti-HBc antibody but none was HBs antigen positive. More numerous subjects had been previously transfused (30%) or had stayed in a country with an endemic virus infection (40%), but of the 6 patients tested for anti-HCV antibodies none was positive. Most patients presented with classical HCC (n = 22), well or moderately differentiated (n = 21) and classified as Edmonson's grade II (Table 2). The nontumorous part of the liver was nonfibrotic in 10 patients (45.5%), fibrotic without septa in 9 (41%), and fibrotic with noncirrhotic septa in 3 (13.5%). Otherwise, 2 young patients (1 man and 1 woman) had the fibrolamellar variant of HCC developed on a nonfibrotic liver. According to selection criteria, control groups did not differ significantly from the HCC/NCL patients with respect to age, sex, and the presence or absence of chronic alcoholism (data not shown). S t u d y o f L i v e r Iron
Iron was found histologically in 83% (20 of 24) of the HCC/NCL samples compared with 37% to 71% in control groups. The difference was significant between HCC/NCL and META and CIR groups. The four noniron-overloaded HCC/NCL patients consisted of three men and one woman ranging in age from 21 to 75 years.
TABLE 2. E d m o n d s o n ' s C l a s s i f i c a t i o n o f HCC, F i b r o s i s S t a g e of t h e N o n t u m o r a l L i v e r , a n d T u m o r S i z e i n 24 P a t i e n t s With HCC Developed on a Noncirrhotic Liver
Edmondson's classification Grade 1 Grade 2 Grade 3 Grade 4 Fibrosis Absent Stage 1 Stage 2 Stage 3 Tumor size (cm)
Classic HCC (n = 22)
Fibrolamellar Variant (n = 2)
1 (4.5%) 20 (90.9%) 1 (4.5%) 0 (0%)
-2 (100%) ---
10 (45.5%) 7 (31.8%) 2 (9.1%) 3 (13.6%) 11.3 _+ 5.7
2 (100%) 0 0 0 12-15
NOTE. Tumor size is expressed as mean +_ SD for patients with classical HCC and as the size of tumor for the two patients with the fibrolamellar variant.
448
TURLIN ET AL
HEPATOLOGY August 1995
2,5
1,5
0,5
Cirrhotic liver
Non-cirrhotic liver
FIG. 1. Hepatic iron index (liver iron concentration/age) in patients with a noncirrhotic liver (HCC/NCL, [ . ] ; META, [[]]; and normal liver [[7]) and in patients with a cirrhotic liver (HCC/CIR, [[]]; and CIR [I]). Results are expressed as means and 95% confidence intervals.
Their mean liver iron concentration was 9.3 _+ 8 #mol/ g. From the 20 iron-overloaded HCC/NCL patients, 18 (90%) presented with parenchymal iron overload. This distribution of iron was significantly much less frequent in controls (HCC/CIR: 54%; CIR: 44%; META: 38%; normal liver: 41%). In the same way, the hepatocytic iron score was significantly higher in the HCC/ NCL patients (11 _+ 8) than in controls, whereas the five groups did not differ significantly according to sinusoidal and portal iron scores (Table 3).
Liver iron concentration was higher in HCC/NCL than in control groups (Table 3). As illustrated in Fig. 1, liver iron concentration/age ratio was significantly increased in HCC/NCL patients (1.4 _+ 1.5) when compared with noncirrhotic controls (normal and metastasis groups) and in HCC/CIR (0.9 + 1.1) when compared to cirrhosis uncomplicated with HCC. Interestingly, the two HCC groups did not differ with respect to the hepatic iron index. The percentage of patients with a hepatic iron index greater than 1 was significantly higher in the HCC/NCL group (63%) than in controls (HCC/CIR: 25%, P < .05; CIR: 8.3%, P < .001; META: 18%, P < .01; normal liver: 21%, P < .01). Four of the 24 HCC/NCL patients (17%) presented with a hepatic iron index > 1.9 compared with 2 in the 88 control patients (2%, P < .05). When excluding these 4 patients, results did not change, except for the hepatic iron index, which became the same for HCC/NCL and normal liver groups (0.9 ___ 0.5 vs. 0.8 + 0.4). DISCUSSION
The major result of the present study is the histological and biochemical demonstration of an unquestionable iron excess in the nontumorous liver of most patients presenting with HCC developed on a noncirrhotic liver and, at a lesser extent, on cirrhosis. Such a finding has not been previously reported by investigators who studied HCC/NCL case series. 2°-23In fact, Melia et al, 2° Kew et al, 22 and Okuda et a123 did not mention the use of Perls' stain in their studies. As for Zillhart et al, 21 they did not mention the results of the iron staining they used. It is likely that these investigators would have paid attention only to severe iron overload compatible with well-defined genetic he-
TABLE 3. Qualitative and Quantitative Iron O v e r l o a d i n HCC/NCL Patients Compared With Control S e r i e s HCC/NCL No. Iron deposits Absent Parenchymal Mesenchymal TIS (Range) HIS SIS PIS LIC (/zmol/g) (Range) LIC/age (HII) (Range) HII < 1 1 < H I I < 1.9 HII -> 1.9
HCC/CIR
CIR
META
Normal Liver
24
20
24
20
24
4 18 2 15.3 ± 11.8 (0-41) 10.6 ± 8.1 3.5 + 2.9 1.1 ± 2.5 81 ± 96 (3-475) 1.4 ± 1.5 (0.07-7.3) 9 11 4
6 7* 6 9.7 ± 9.9* (0-28) 6.3 ± 6.2* 2.6 + 3 0.8 _+ 1.4 52 _+ 67* (4-315) 0.9 ± 1.1 (0.1-4.9) 15" 4 1
15 4* 5 3.2 _+ 5.35 (0-19) 2.1 ± 3.55 1 x 1.9t 0.04 _+ 0.2t 21 x 215 (5-55) 0.4 ± 0.35 (0.2-1.5) 22t 2 0
12 3t 5 2.5 _+ 3.75 (0-12) 1.4 ± 2.35 1.2 _+ 1.9t 0t 36 ± 22* (13-106) 0.6 ± 0.3# (0.08-1.4) 17t 3 0
7 7t 10 8.3 _+ 4.9* (0-19) 5.8 _+ 3.6* 2.2 ± 1.9 0.2 ± 0.5 41 _+ 20* (18-114) 0.8 _+ 0.4* (0.2-1.9) 19t 4 1
Abbreviations: TIS, total iron score; HIS, hepatocytic iron score; SIS, sinusoidal iron score; PIS, portal iron score; LIC, liver iron concentration; HII, hepatic iron index. * P < .05. t P < .01. $ P < .001 compared with HCC/NCL.
HEPATOLOGYVol. 22, No. 2, 1995
mochromatosis. In our study, except for four HCC/NCL patients, most cases presented with mild iron excess and might have been undiagnosed on routine examination. It is noteworthy that iron excess in HCC patients was both quantitatively and qualitatively different from that of control subjects. Indeed, in 75% of HCC/ NCL cases and in 35% of HCC/CIR, iron overload was mainly located within periportal (or perinodular) parenchymal cells. On the other hand, in non-HCC patients, iron excess, when found, was mainly located within mesenchymal cells without any lobular predominance. In addition, the hepatic iron index was significantly greater in HCC/NCL and in HCC/CIR patients than it was in their respective control groups. These results suggest that, in the noncirrhotic liver and, to a lesser extent, in the cirrhotic liver, there is a relationship between HCC and the finding of a parenchymal iron excess. Only homozygous genetic hemochromatosis is known to be responsible for both iron overload and HCC in patients with cirrhosis as well as in noncirrhotic patients. 6 However, when based on usual diagnostic criteria, i.e., a hepatic iron index - 1.9 in the absence of any other cause of iron overload, 16 an underlying homozygous genetic hemochromatosis could be considered only in five of our patients with HCC (four HCC/NCL and one HCC/CIR subjects with a hepatic iron index ranging from 1.9 to 7.3). Moreover, a heterozygous genetic hemochromatosis could be discussed in the other HCC cases, but no firm conclusion could be drawn with respect to the putative genetic status of these patients. Other causes of iron overload such as repeated transfusions, patent dysmyelopoiesis, long-term iron intake, and chronic alcoholism were ruled out. For the latter, it was unlikely that an excessive alcohol consumption could be responsible for the iron excess found in our patients because control groups, matched according to the presence or absence of chronic alcoholism, presented with a significantly lower liver iron level. Finally, elevated serum iron and ferritin values have been reported in patients with HBV and HCV infections, but, to our knowledge, no significant histological or biochemical liver iron overload has been documented in such conditions. 24 Then, on the contrary, the role of HCC in the development of liver iron excess could be suggested but: (1) redistribution of iron because of the presence of a space-occupying lesion was unlikely because no significant iron overload was found in the metastasis group and because we failed to find any correlation between tumor size and liver iron concentration; and (2) the concept of a metabolic "paraneoplastic" liver iron overload was not addressed because the course of iron overload after the removal of the tumor remained unknown. Whatever the putative causes of HCC and iron excess in this series, it could be suggested that tumor development might be triggered by an associated liver iron overload. Indeed, a direct carcinogenic effect of iron has not been documented in animals z~ as well as in humans, 262s but iron has been shown to act as a cocar-
TURLIN ET AL
449
cinogen factor in tumor cell growth in vitro and in
vivo.IO,12,13
CONCLUSION
The contribution of the present study is the showing of a mild but unquestionable parenchymal iron excess in the nontumorous liver of most patients presenting with HCC developed on a noncirrhotic liver and, at a lesser extent, on cirrhosis. The pathophysiology of this iron overload remains unexplained and deserves further h u m a n and experimental studies. These results (1) should lead to study of the putative role of iron in the development of HCC in both the cirrhotic and the noncirrhotic patient, whatever the identification of an underlying liver disease, and (2) provide further argument to take into account and to treat any liver iron excess, even when mild.
Acknowledgment: We thank Professor Yoichiro Kondo for his kind reassessment of the HCC/NCL cases, and Jean-Yves Robert and R~mi Trinquart for their excellent technical assistance. REFERENCES 1. Antoine D, Braun P, Cervoni P, Schwartz P, Lamy P. Le cancer bronchique des mineurs de fer de Lorraine peut-il ~tre consid~r~ comme une maladie professionnelle?Rev Fr Mal Resp 1979; 7:6365. 2. Stevens R, Beasley R, Blumberg B. Iron-binding proteins and risk of cancer in Taiwan. J Natl Cancer Inst 1986;76:605-610. 3. Blumberg R, Chopra S, Ibrahim R, Crawford J, Farraye F, Zeldis J, Berman M. Primary hepatocellular cancer in idiopathic hemochromatosis after reversal of cirrhosis. Gastroenterology 1988;95:1399-1402. 4. Fellows I, Stewart M, Jeffcoate W, Smith P, Toghill P. Hepatocellular carcinoma in primary h~emochromatosis in the absence of cirrhosis. Gut 1988;29:1603-1606. 5. Weinberg E. Roles of iron in neoplasia: promotion, prevention and therapy. Biol Trace Elem Res 1992;34:123-140. 6. Deugnier Y, Guyader D, Crantock L, Lopez J, Turlin B, Yaouanq J, Jouanolle H, et al. Primary liver cancer in genetic hemochromatosis: a clinical, pathological and pathogenetic study of 54 cases. Gastroenterology 1993;104:228-234. 7. Weinberg E. Association of iron with respiratory tract neoplasia. J Trace Elem Exp Med 1993;6:117-123, 8. Nelson R, Davis F, Sutter E, Sobin L, Kikendall W, Bowen P. Body iron stores and risk of colonic neoplasia. J Natl Cancer Inst 1994;86:455-460. 9. Fernandez-Pol J. Iron: possible cause of the G1 arrest induced in NRK cells by picolinic acid. Biochem Biophys Res Commun 1977;78:136-143. 10. Bergeron R, Streiff R, Elliott G. Influence of iron on in vivo proliferation and lethality of L1210 cells. J Nutr 1985; 115:369374. 11. Harm H, Stahlhut M, Hann C. Effect of iron and desferoxamine on cell growth and in vitro ferritin synthesis in human hepatoma cell lines. HEPATOLOGY1990; 11:566-569. 12. Thompson H, Kennedy K, Witt M, Juzefyk J. Effect of dietary iron deficiency or excess on the induction of mammary carcinogenesis by 1-methyl-l-nitrosourea. Carcinogenesis 1991; 12:111114. 13. Hann H, Stahlhut M, Rubin R, Maddrey W. Antitumor effect of desferoxamine on human hepatocellular carcinoma growing in athymic nude mice. Cancer 1992;70:2051-2056. 14. Barry M, Sherlock S. Measurement of liver-iron concentration in needle-biopsy specimens. Lancet 1971;2:100-103. 15. Olynyk J, O'Neil R, Britton R, Bacon B. Determination of hepatic
450
16. 17. 18. 19.
20. 21.
22.
TURLIN ET AL iron concentration in fresh and paraffin-embedded tissue: diagnostic implications. Gastroenterology 1994; 106:674-677. Bassett M, Halliday J, Powell L. Value of hepatic iron measurements in early hemochromatosis and determination of the critical iron associated with fibrosis. HEPATOLOGY1986;6:24-29. Edmondson H, Steiner P. Primary carcinoma of the liver: a study of 100 cases among 48900 necropsies. Cancer 1954;7:462-503. Anthony P, Ishak K, Nayak N, Poulsen H, Scheuer P, Sobin L. The morphology of cirrhosis. J Clin Pathol 1978; 31:395-414. Deugnier Y, Lor~al O, Turlin B, Guyader D, Jouanolle H, Moirand R, Jacquelinet C et al. Liver pathology in genetic hemochromatosis: a review of 135 homozygous cases and their bioclinical correlations. Gastroenterology 1992;102:2050-2059. Melia W, Wilkinson M, Portmann B, Johnson P, Williams R. Hepatocellular carcinoma in non cirrhotic liver: a comparison with that complicating cirrhosis. Quart J Med 1984; 211:391-400. Zillhart P, Szekely A, Houssin D, Santelli G, Lemaigre G, Bismuth H. Carcinome h~patocellulaire sur foie non cirrhotique. Etude anatomo-clinique de 22 cas d'h~patectomies. Facteurs histo-pronostiques. Ann Pathol 1985;5:239-247. Kew M. Hepatocellular carcinoma with and without cirrhosis. A
HEPATOLOGYAugust 1995
23. 24. 25. 26. 27.
28.
comparison in southern African blacks. Gastroenterology 1989;97:136-139. Okuda K, Nakashima T, Kojiro M, Kondo Y, Wada K. Hepatocellular carcinoma without cirrhosis in Japanese patients. Gastroenterology 1989;97:140-146. Di Bisceglie A, Axiotis C, Hoffnagle J, Bacon B. Measurements of iron status in patients with chronic hepatitis. Gastroenterology 1992; 102:2108-2113. Park C, Bacon B, Brittenham E, Tavill A. Pathology of dietary carbonyl iron overload. Lab Invest 1987;57:555-563. Adams P, Speechley M, Kertesz A. Long-term survival analysis in hereditary hemochromatosis. Gastroenterology 1991; 101:368372. Fargion S, Mandelli C, Piperno A, Cesana B, Fracanzani A, Fraquelli M, Bianchi P e t al. Survival and prognostic factors in 212 Italian patients with genetic hemochromatosis. HEPATOLOGY 1992; 15:655-659. Colombo M, de Franchis R, Del Ninno E, Sangiovanni A, de Fazio C, Tommasini M, Donato M, et al. Hepatocellular carcinoma in Italian patients with cirrhosis. N Engl J Med 1991; 325:675-680.