Correspondence
Hepatocellular Carcinoma
N Engl J Med 1993; 328:64-65January 7, 1993
- Article
To the Editor:
Colombo et al. showed (Sept. 5, 1991, issue)1 that yearly screening for hepatocellular carcinoma with an alpha-fetoprotein assay and real-time ultrasonography failed to improve the rate of detection of small and potentially operable tumors in Italian patients with cirrhosis. The authors suggested that more frequent screening (i.e., every four to six months) might be more effective.
From July 1987 to September 1989, 193 patients with cirrhosis (127 men and 66 women; mean age, 56 years) underwent screening for hepatocellular carcinoma at our institution. Clinical evaluation, serum alpha-fetoprotein measurements, and real-time ultrasonography were performed every six months. The causes of cirrhosis were alcoholism (76 percent), hepatitis B (5 percent), hepatitis C (2.6 percent), and miscellaneous factors (16.4 percent). The diagnosis of hepatocellular carcinoma was either made by biopsy or based on the association of a serum alpha-fetoprotein level above 500 μg per liter and a focal lesion on ultrasonography. Hepatocellular carcinoma was discovered in 7 of the 193 patients (at enrollment in 3 patients, at 12 months in 3, and at 18 months in 1), for an overall incidence of 1.8 percent per year. The tumor was unifocal in three patients (only one tumor was less than 3 cm in diameter) and multifocal in the remaining four patients. Only one patient had liver transplantation. The probability of survival was 60 percent at 6 months and 30 percent at 12 months. Of the 186 patients in whom hepatocellular carcinoma did not develop, 16 required further investigations because of a focal lesion on ultrasound scanning or elevation of the serum alpha-fetoprotein level (or both).
During the same period, 54 patients with cirrhosis (42 men and 12 women; mean age, 62 years) were admitted to our institution because of symptomatic hepatocellular carcinoma. None of these patients had been screened for the disease previously. The causes of cirrhosis in this group were alcoholism (78 percent), hepatitis B (15 percent), and miscellaneous factors (7 percent). Of these 54 patients, 9 had a unifocal tumor less than 3 cm in diameter, 10 had a unifocal tumor more than 3 cm in diameter, and 35 had multifocal tumors. Two patients underwent liver transplantation. The probability of survival was 43 percent at 6 months and 35 percent at 12 months. The patients who had been screened for hepatocellular carcinoma and the patients who had been admitted for symptomatic hepatocellular carcinoma were comparable in age, sex, and cause of cirrhosis. The two groups did not differ significantly in the rate of unifocal tumors less than 3 cm in diameter or in the probability of survival.
Our data confirm the poor results reported by Colombo et al. in European patients, and contrast with the encouraging results of studies carried out in Asia2,3. Furthermore, our data suggest that intensifying screening by performing investigations every six months increases the cost but may not improve the rate of detection of small tumors.
3 ReferencesFrancois Durand, M.D.
Catherine Buffet, M.D.
Gilles Pelletier, M.D.
Herve Hagege, M.D.
Olivier Ink, M.D.
Jean-Pierre Etienne, M.D.
Hopital Bicetre, 94275 Le Kremlin Bicetre, CEDEX France1
Colombo M ,de Franchis R ,Del Ninno E , et al. Hepatocellular carcinoma in Italian patients with cirrhosis. N Engl J Med 1991;325:675-680
Full Text | Web of Science | Medline2
Sheu JC ,Sung JL ,Chen DS , et al. Early detection of hepatocellular carcinoma by real-time ultrasonography: a prospective study. Cancer 1985;56:660-666
CrossRef | Web of Science | Medline3
Kobayashi K ,Sugimoto T ,Makino H , et al. Screening methods for early detection of hepatocellular carcinoma. Hepatology 1985;5:1100-1105
CrossRef | Web of Science | Medline
To the Editor:
The article by Colombo et al. concluded that screening for hepatocellular carcinoma in Italian patients with cirrhosis did not improve outcome, whereas studies from Asia had reported improved survival after screening.
In Asia, screening procedures are performed frequently. Recent studies from two centers in Japan describe surveillance protocols that include ultrasound examinations conducted every three months1,2. In contrast, ultrasound examinations were performed yearly in the patients of Colombo et al. except in the minority with alpha-fetoprotein concentrations above 400 μg per liter. It is possible that more curable cases of hepatocellular carcinoma might have been diagnosed if examinations had been conducted at shorter intervals, particularly since 65 percent of the tumors detected one year after normal ultrasound examinations were multifocal or more than 3 cm in size.
We wish to point out various options relevant to the diagnosis, staging, and treatment of this disorder. Computed tomography (CT) arterial portography (during infusion of a contrast agent into the superior mesenteric artery), as well as CT after infusion of iodized oil with avidity for hepatocellular carcinoma (Lipiodol) into the hepatic artery, are superior to both conventional CT scanning and ultrasound examination for the detection of small hepatocellular carcinomas3. Magnetic resonance imaging (MRI) has also been reported to be superior to CT scanning and ultrasound examination in the detection of liver tumors4,5. These techniques may also be more sensitive than conventional CT scanning for staging. However, considerations such as cost and availability need to be balanced against the possibility of improved outcome.
Colombo et al. note that six of their patients had a “clear-cut diagnosis of an angioma” on ultrasound examination. Although this did not affect the authors' results, it is useful to remind readers that the diagnosis of hemangioma by ultrasound examination is far from 100 percent specific, particularly in patients at high risk6. In one series, 37 percent of hepatocellular carcinomas were echogenic,7 and small hepatocellular carcinomas can mimic hemangioma8. Accordingly, many radiologic authorities recommend a more specific confirmatory test, such as MRI or scanning with labeled red cells6.
Therapy now available includes transarterial chemo-embolization, and immunotherapy is under investigation9. Percutaneous intralesional instillation of absolute ethanol under ultrasound guidance has a reported four-year survival of 79 percent10. Groups of Japanese investigators have found that this technique yields better survival than surgery, even for resectable lesions8,10.
Despite the pessimistic outlook voiced in this small study, the availability of increasingly accurate methods for the detection of hepatocellular carcinoma and effective nonsurgical therapy justify optimism that aggressive screening of high-risk patients worldwide will be beneficial.
10 ReferencesRonald H. Wachsberg, M.D.
Donald G. Mitchell, M.D.
Thomas Jefferson University Hospital, Philadelphia, PA 191071
Oka H ,Kurioka N ,Kim K , et al. Prospective study of early detection of hepatocellular carcinoma in patients with cirrhosis. Hepatology 1990;12:680-687
CrossRef | Web of Science | Medline2
Okazaki N ,Yoshino M ,Yoshida T , et al. Early diagnosis of hepatocellular carcinoma. Hepatogastroenterology 1990;37:480-483
Web of Science | Medline3
Takayasu K ,Moriyama N ,Muramatsu Y , et al. The diagnosis of small hepatocellular carcinomas: efficacy of various imaging procedures in 100 patients. AJR Am J Roentgenol 1990;155:49-54
Web of Science | Medline4
Curati WL ,Halevy A ,Gibson RN ,Carr DH ,Blumgart LH ,Steiner RE . Ultrasound, CT, and MRI comparison in primary and secondary tumors of the liver. Gastrointest Radiol 1988;13:123-128
CrossRef | Medline5
Stark DD ,Wittenberg J ,Butch RJ ,Ferrucci JT Jr . Hepatic metastases: randomized, controlled comparison of detection with MR imaging and CT. Radiology 1987;165:399-406
Web of Science | Medline6
Nelson RC ,Chezmar JL . Diagnostic approach to hepatic hemangiomas. Radiology 1990;176:11-13
Web of Science | Medline7
Maringhini A ,Cottone M ,Sciarrino E , et al. Ultrasonography and alpha-fetoprotein in diagnosis of hepatocellular carcinoma in cirrhosis. Dig Dis Sci 1988;33:47-51
CrossRef | Web of Science | Medline8
Okuda K . Diagnosis and nonsurgical treatment of hepatocellular carcinoma. Hepatogastroenterology 1990;37:159-164
Web of Science | Medline9
The Liver Cancer Study Group of Japan
Web of Science | Medline10
Ebara M ,Nihei T ,Ohto M . Intratumoral injection of absolute ethanol for treatment of small hepatocellular carcinoma. Naika 1988;61:665-669
To the Editor:
Wands and Blum1 state in their excellent editorial on the report by Colombo et al. that cirrhosis associated with genetic diseases such as hemochromatosis, alpha1-antitrypsin deficiency, and tyrosinemia may culminate in liver cancer. To this list should be added porphyria, an important cause of hepatocellular carcinoma2-4.
Increased porphyrin biosynthesis and underutilization of porphyrin are the mechanisms responsible for accumulation of porphyrin in the liver due to N-methyl protoporphyrin and aminoevulinate synthase.
4 ReferencesSaeed Ahmad, M.D., F.R.C.P.
1000 Brookside Dr., Fairmont, WV 265541
Wands JR ,Blum HE . Primary hepatocellular carcinoma. N Engl J Med 1991;325:729-731
Full Text | Web of Science | Medline2
Ahmad S . Porphyria and liver cancer. Am J Med 1991;91:102-103
CrossRef | Web of Science | Medline3
Litner F ,Wetterberg L . Hepatocellular carcinoma in patients with acute intermittent porphyria. Acta Med Scand 1984;215:271-274
CrossRef | Web of Science | Medline4
Kauppinen R ,Mustajoki P . Acute hepatic porphyria and hepatocellular carcinoma. Br J Cancer 1988;57:117-120
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: We appreciate the comments of Drs. Wachsberg and Mitchell about the low rate of detection of potentially curable hepatocellular carcinomas in our patients with cirrhosis. They underline an issue already acknowledged in our paper -- that better results might have been obtained if we had performed more frequent ultrasound examinations in patients with normal levels of alpha-fetoprotein. It should be noted that the cutoff level for alpha-fetoprotein, above which patients were enrolled in the more aggressive follow-up program -- i.e., quarterly ultrasound scanning -- was 20 μg per liter, and not 400 μg per liter as Wachsberg and Mitchell stated. Nevertheless, it is true that more than 300 patients had yearly ultrasound scans. We can only repeat what we concluded in our study: that quarterly ultrasound scanning in all patients would involve substantial increases in costs owing to the low incidence of newly diagnosed tumors in patients with cirrhosis. This is even more true of the alternative techniques mentioned by Wachsberg and Mitchell. In fact, conventional CT scanning, CT arterial portography, CT after infusion of hepatocellular carcinoma-avid Lipiodol into the hepatic artery, and MRI are all more expensive, less widely available, and except for MRI, more invasive than ultrasound scanning. Therefore, even if some of the above techniques are substantially superior to conventional ultrasound scanning in detecting hepatocellular carcinoma, their cost alone will probably prevent them from replacing ultrasound scanning for screening. As for the comment on the diagnosis of hemangioma by ultrasound examination, we agree that a number of small hepatocellular carcinomas may mimic hemangioma. However, this did not happen in our series, since none of the patients whose initial diagnosis was hemangioma turned out to have hepatocellular carcinoma during a median follow-up period of 33 months.
As far as the letter of Dr. Durand and colleagues is concerned, we are pleased to see results similar to ours from a different Western country, and even more so because those findings were made in a patient population in which alcoholism was the leading cause of cirrhosis, as compared with viral hepatitis in our patients.
M. Colombo, M.D.
R. de Franchis, M.D.
Institute of Internal Medicine, University of Milan, 20122 Milan, Italy- Citing Articles (2)
