March 2001, Volume 51, Issue 3

Short Reports

Lack of Correlation between Aipha-Fetoprotein and Tumor size in Hepatocellular Carcinoma

S. Sharieff  ( Department of Medicine, The Aga Khan University Hospital. Karachi. )
I. A. Burney  ( Department of Medicine, The Aga Khan University Hospital. Karachi. )
A. Salani  ( Department of Medicine, The Aga Khan University Hospital. Karachi. )
T. Siddiqui  ( Department of Medicine, The Aga Khan University Hospital. Karachi. )


Objective: To determine whether there was a correlation between tumor size and alpha feto-protein (a-FP) levels in hepatocellular carcinoma.
Setting: Retrospective study in tertiary referral center with specialist Oncology services in Southern Pakistan. Subjects: Consecutive patients with biopsy-proven hepatocellular carcinoma diagnosed between January 1994 and June 1998.
Main Outcome Measures: Correlation between a-FP levels and maximum tumor diameter.
Results: The mean tumor size was 8.3 ± 4.2 cm. The mean (J-FP level was 17.027 ng/ml. Twenty four percent patients had an u-FP level which was within the normal limits (<10 ng/ml). There was no correlation between tumor size and a-FP levels (r = -0.155; p = 0.129).
Conclusion: There was no correlation between the tumor size and a-FP levels (JPMA 51:123;2001).


Hepatocellular Carcinoma (HCC) is an uniformly fatal malignancy with a higher incidence in certain parts of Asia and Africa1-3. The only curative modality of treatment is resection, however, the vast majority of cases are detected at a stage when the tumor is unresectable. Although there are no satisfactory screening procedures for early detection, serum alpha fetoprotein (a-FP) and ultrasound of abdomen have been used either singly or in combination, but only with a modest degree of success
Studies have shown variable a-FP in patients with HCC at the time of diagnosis6-9, However, the majority of these studies had a small sample size. We conducted a retrospective analysis to see whether a correlation existed between the tumor size and the level of cx-FP in a large population of patients diagnosed to have HCC and hence to see if a-FP could be used for screening and early detection of HCC in our group of patients.

Patients, Methods and Results

The data of consecutive patients diagnosed to have HCC and admitted to the hospital between January 1994 to June 1998 were analyzed retrospectively. Quantitative measurement of a-FP was done on an immulite-automated analyzer. The tumor size measurement was available in two dimensions on ultra sound examination. For purposes of analysis, the largest diameter was used. A Pearson correlation was applied to see the association between tumor size and level of a-FP. Analysis of variance (ANOVA) was applied to study the association between the survival time and levels of a-FP. A p-value of < 0.05 was taken as a statistically significant.
A total of 201 patients were admitted with a diagnosis of HCC over the period of 4-l/2years. There were 149 males and 52 females. Although the mean ci-FP was 17,027 ng/ml, but there was a marked variation in its level (Figure 1).

Twenty four percent of the patients had a normal a-FP (<10ng/ml) level, while 35% had values more than 1000 ng/ml.
The mean tumor size was 8.3 cm ± 4.0 cm. The majority of patients (79.49%) had tumor size more than 5 cm in diameter.

Figure 2 clearly shows a lack of correlation between the level of a-FP and the tumor size.
We also compared the survival rates with a-FP levels and found that the mean survival time was 1 7.6 ± 2 1I .5 weeks for patients with a-FP level less than 10 ng/ml, 17.9 ± 21.7 weeks for patients with a-FP levels between 10 to 1000 ng/ml and 13.2 ± 15.6 weeks for patients with a-FP level greater than 1000 ng/ml. The mean survival time was not statistically significantly different with different levels of a-FP ( p-value =0.352). Also, there was no significant difference in the level of ct-FP between patients who had HBsAg and Anti-HCV related HCC. (13,395 ±, 31,211 ng/ml and 5,938 ± 18,737 ng/ml respectively, p-value 0.287).


a-FP was elevated in 76% of cases with Hcc. However, there was no correlation between the tumor size and alpha-fetoprotein levels and overall survival, in a large group of patients. This observation is consistent with some previous reports1.


1.Sato A, Kato Y, Nakata K, et al. Aetiopathogenesis of hepato cellular carcinoma. J. Gastroenterol. Hepatol., 1996: 11 944 - 948.
2.Okuda K. Early recognition of hepatocellular carcinoma. llcpatology, 1986: 6:729-738.
3.Bosch F.X, Munoz N. Hepatocellular carcinoma in the world Epidemiological question. In: Tabor E. Di Bisceglie AM. Purcell RU, eds. Epidemiology. Pathology, and Treatment of 11CC in North America, Houston, Gulf, 1991: pp. 35-54.
4.Adrian M, Di Bisccglie A.M. Hepatitis C and Hepatocellular Carcinoma. flepatologv. 1997; 26:34s - 38s.
5.Paterson D , Ganne N. Trinchet J.C et al. Prospective study of screening for bepato cellular carcinoma in Caucasian patients with cirrhosis, J. Hepatol.,1994: 20:65-71.
6.Tanq NA. Review of fitly cases of Hepatocellular carcinoma. Pak. J. Med. Res., 1990: 29:97 - 99.
7.Zuberi Si, Lodhi T.Z, Nazli A. Alpha - Fetoprotein in-patients with primary liver cancer and other liver disease. J Pak Med Assoc. 1977: 27:348 - 350.
8.Ahmad W, Lodhi T.Z, Zuberi S.J. Alpha fetoprotien in liver disease. i Pak Med Assoc. 1988: 38:234 - 236.
9.Bolondi L, Gaiani S, Benzi G. et al. Ultrasonography and guided biopsy in the diagnosis of Hepatocellular carcinoma. ltal. J. Gastroentrol., 1992: 24:46-49.
10.Nakao A, Taniguchi K, lnoue 5. et al. Usefulness of simultaneous detennination of alpha-fetoprotein and Des, gamma carboy prothrombin in Hepatocellular carcinoma, Semm. Surg. Oncol., 1996: 12:160 - 63.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: