Proceedings of the American Thoracic Society

Viruses have been feared as major pathologic agents, a scourge of humans and other species. However, they have a more textured relation to their hosts. This article focuses on hepatitis B virus (HBV), one of the most common and deadly agents that infect humans and that has a powerful selective effect on survival, which also has complex and subtle interactions with its host that are not pathologic. HBV affects populations and their evolution in an imaginative manner that is a model for understanding how other microorganisms can both plague and assist their hosts at different times and in different environmental and genomic contexts. The host responses to infection with HBV are affected, in large part, by a series of polymorphic alleles at several loci on the human genome, and these loci are also related to other infections. HBV has an effect on the sex ratio in populations, both at birth and later in life. The effects of these dynamic systems on the demography, population biology, and microevolution can be profound.

The most common outcome after infection is the development of antibody against the surface antigen of HBV (Table 1). The infected person experiences no symptoms and ordinarily has lifelong protection against further infection. A significant number of those infected will develop acute hepatitis. This is ordinarily a self-limited disease that includes flulike symptoms, fever, joint pains, malaise, severe loss of appetite, and, in time, jaundice. Most patients recover in several weeks or months with no further symptoms, but a small percentage may develop fulminant hepatitis, which has a high mortality rate. About 10% of patients with acute hepatitis will retain the infection and develop chronic hepatitis. They can remain ill for months or years and this may result in liver dysfunction, liver failure, and early death. Some of those chronically infected individuals may proceed to primary cancer of the liver, which is difficult to treat and life shortening. In some cases, chronic liver disease may occur without an antecedent acute phase.


1. Anti-Hbs. Development of antibody against the surface antigen.
 Usually protects against further disease.
2. Acute hepatitis. Usually a self-limited disease with complete recovery.
 Some may result in fulminant hepatitis, which has a high mortality rate.
 Some cases proceed to chronic HBV infection with increased risk of chronic liver disease and primary cancer of the liver.
3. The carrier state. Chronic infection with HBV that is asymptomatic for decades.
 At increased risk for chronic liver disease and primary cancer of the liver.
4. Chronic liver disease
5. Primary cancer of the liver
6. Other serologic reactions, i.e., anti-HBc, anti-HBx, anti-HBe, HBeAg, HBV DNA, etc., that denote different stages of disease and infection

Definition of abbreviation: HBV = hepatitis B virus.

Many infected people, particularly those infected at or shortly after birth, will develop the HBV carrier state. The percentage of those infected who become carriers varies greatly from population to population. For example, in China and elsewhere in East Asia, 5 to 20% of the population may be carriers, whereas in the United States and northern Europe, the prevalence is less than 1%. Carriers can remain asymptomatic for years and decades but are at greatly increased risk for the development of chronic liver disease and primary cancer of the liver.

Primary cancer of the liver is one of the most common cancers in the world, and particularly so in sub-Saharan Africa, China, and East Asia. It is the third most common cause of death from cancer in males and the seventh most common cause in females. Its toll is about one million deaths a year. HBV is believed to cause about 70 to 85% of the cases; many of the rest are caused by hepatitis C virus (HCV). Other factors contribute to the etiology of primary cancer of the liver, including liver toxins such as alcohol. Treatments are available, but they are far from ideal.

There are other hepatitis viruses and they are shown Table 2. They are different from each other and several are in separate families. Although the clinical manifestations may be similar, the epidemiology, method of spread, chronicity, and other features are different. Hepatitis A virus is transmitted by the fecal–oral route and is rarely chronic. As noted, HBV causes acute and chronic disease and the carrier state. It is transmitted from infected mothers to their children, by sexual transmission, and by transfer of blood from one person to another (as in blood transfusion). HCV is transmitted in the same way as HBV, but it is less infectious. Hepatitis D virus is an unusual, very small virus that is “wrapped” in the surface antigen of HBV. It can only infect people who are already infected with HBV, or who are infected at the same time with both viruses. Hepatitis E virus is transmitted in the same manner as hepatitis A virus and can occur in large and devastating epidemics.



Size (kb)
HAVRNA7.5Picornavirus (hepatovirus)
HCVRNA9.4+Pestivirus-or flavivirus-like

Claicivirus-like or alpha- like supergroup

Definition of abbreviations: HAV = hepatitis A virus; HBV = hepatitis B virus; HCV = hepatitis C virus; HDV = hepatitis D virus; HEV = hepatitis E virus.

The focus of this article is on HBV.

It is difficult to estimate the total number of deaths due to infection with HBV; an estimate of over a million is often used. An estimate of the expected number of deaths due to prevalent HBV carriers compared with deaths due to prevalent cases of HIV was made in 1999. (The number of HIV prevalent infections has increased since that time.) Using a high estimate for deaths due to HBV, the total expected deaths is 92.8 million; using a low estimate, the number is 55.7 million. The equivalent number for deaths associated with HIV was somewhat less; estimates for HIV deaths in 2005 would be higher.

It is estimated that two billion people alive today, one-third of the world's population, have been infected with HBV. Most of these would have developed anti-HBs and remained well, but many have developed clinical hepatitis and some 375 million worldwide are currently carriers of HBV. Humans now have and have had in the past a highly significant interaction with HBV. Mortality by sex, age, and carrier status in Haimen City, China, is shown in Figure 1. Male carriers have a higher mortality than noncarrier males, and female carriers have a higher mortality than noncarrier females. Male carriers and noncarriers have a much higher mortality than females. These differences increase with age. Therefore, HBV kills more males, particularly older males, than females. An effective vaccination program (see below) will increase the male-to-female sex ratio, especially in the older age groups. As shown later, chronic infection with HBV appears to have the opposite effect in newborns. The worldwide distribution of the approximately 375 million carriers of HBV is shown in Figure 2. The prevalence is extremely high in Asia, the Pacific, Africa (particularly sub-Saharan Africa), in the Amazon River basin of South America, and, strangely, among the Inuits of the Arctic.

The research that resulted in the discovery of HBV and the invention of the HBV vaccine began as a basic, nondirected study of human genetic polymorphisms and differential susceptibility to disease. My colleagues and I did not start off to discover the virus; as is often the case in science, during the course of this “pure,” nonapplied research, the path we were following led to an unexpected and practical outcome. In the search for antigenic polymorphic variation in human serum proteins, we used the serum of patients who had received many transfusions. Our hypothesis was that transfused patients would develop antibodies against an antigenic protein variant inherited by the donor and present in his or her blood but that had not been inherited by the recipient of the blood. Using this method, we found a complex and medically important polymorphism of the serum lipoproteins. When we continued the studies, we found another antibody, which we subsequently found to be directed to the outer coat of the HBV. In time, this became a widely used clinical laboratory test that could detect HBV in blood donors. Widespread use of this test soon eliminated post-transfusion hepatitis due to HBV. Subsequently, a test for HCV in the blood was introduced that further decreased the incidence of post-transfusion hepatitis.

The electron microscope image shows the three particles that constitute the HBV complex (Figure 3). The large 42-nm particle is the whole virus. It includes the enveloping surface antigen (HBsAg) and the core antigen (HBcAg) that surrounds the circular, partially double-stranded DNA. HBV is an extremely small virus—smaller than polio virus—with only 3,200 base pairs. In addition to the 42-nm whole virus particles that are pathogenic and infectious, there are small, circular particles that are 22 nm in circumference, and elongated particles of the same diameter and of varying length. They consist only of the surface antigen HBsAg and are not infectious or pathogenic. In 1969, my colleague Irving Millman and I invented a vaccine that was made by separating the small HBsAg particles from the whole virus particles, discarding the infectious virus, treating the resulting surface antigen particles to kill any remaining virus, and adding adjuvants as necessary. This was a unique method of manufacturing vaccine that had not been used before or since. This vaccine was field tested, approved by the U.S. Food and Drug Administration (FDA) for safety and efficacy, and has been used in tens of millions of people.

The field trials were conducted by Dr. Wolf Szmuness and his colleagues (Szmuness W, Stevens CE, Zang EA, Harley EJ, Kellner A. A controlled clinical trial of the efficacy of the hepatitis B vaccine [Heptavax B]: a final report. Hepatology 1981;1: 377–385) in a U.S. population at very high risk for infection with HBV. They found the vaccine to be highly effective (> 90%) in preventing infection and with no evidence of toxicity. The viral DNA consists of four reading frames: S, which produces HBsAg; C, which produces HBcAg; X, which produces HBxAg involved in replication and carcinogenesis; and P, which is responsible for the DNA polymerase required for viral replication and reverse transcriptase that facilitates the insertion of the HBV sequences into the human host genome (Figure 4). The S reading frame has multiple start codons that result in S proteins of different sizes. In the 1980s, several groups produced a recombinant S-antigen vaccine by incorporating the S reading frame, or parts of it, in yeast and other cells. This was also field tested and found to be effective and safe. Most HBV vaccine is now made by the recombinant method. It was the first widely used human recombinant vaccine and is, I believe, the only recombinant human vaccine now in common use.

National vaccination programs have been in place since the early 1980s. The effect on the prevalence of HBV carriers and the incidence of acute hepatitis is impressive. In a regional study in China, the prevalence of HBV carriers dropped from 16 to 1.4%. In the United States, new infections with HBV have dropped from 260,000 in the 1980s to about 78,000 in 2001. In Alaska, acute hepatitis B infection dropped from 215 cases per 100,000 population before the vaccination program to 7 to 14 cases in 1993 after the program was in place. In 1995, no cases were reported. In Gambia, West Africa, the prevalence of chronic infection in the young had dropped from 10.0 to 0.6%.

In Afragola, Italy, a community with very high rates of HBV infection and of morbidity and mortality from liver disease, the prevalence of HBsAg in males up to 12 yr old dropped from 10.5% in 1978 to 0.8% in 1993 after the vaccination program was in place for 10 yr. The prevalence of anti-HBc dropped from 52.6 to 1.2%. There was also a drop in the prevalence of HBsAg in the mostly unvaccinated population—that is, males aged 13 to 60 yr. HBsAg dropped from 18% in 1978 to 5.5% in 1989. This implies that the reduction of the carrier prevalence in the vaccinated group has an indirect effect on unvaccinated carriers and susceptible individuals, and suggests that there may an amplification characteristic to the vaccination program. Similar observations have been made in other population-based studies.

Perhaps the most remarkable effect of the program has been in the decrease in primary cancer of the liver in vaccine-impacted populations. In Taiwan, the average annual incidence in the age range of 6 to 14 yr fell from 0.7/100,000 in 1981–1986 (before vaccination) to 0.36/100,000 in 1990–1994 (after vaccination). The annual incidence in children aged 6 to 9 yr fell from 0.52/100,000 in those born in 1974–1984 (pre–vaccination program) to 0.13/100,00 in those born in 1986–1988 (post–vaccination program). In South Korea, the study cohort included 370,285 males aged 30 yr or older. Of these, 18,914 (5%) were HBsAg positive and 78,094 (21%) anti-HBs positive. A total of 273,277 had no markers. Of these, 35,934 (13.2%) were vaccinated during 1985. The prevalence of primary cancer of the liver was determined from 1986 to 1989. The unvaccinated and chronic carriers of HBsAg had an incidence of 18.1 per 100,000 population. The incidence of cancer of the liver in those who were vaccinated dropped to 0.58 and that in those with “natural” anti-HBs dropped to 0.34.

In both of these studies, the drop was dramatic. If these findings are confirmed in subsequent studies, the HBV vaccine is the first preventive cancer vaccine and the vaccination program is the major medical intervention program for the prevention of cancer. These results give hope that other cancer vaccination programs will be instituted. The recent announcement of the successful field trial of a vaccine to prevent papilloma virus infection—a major cause of cancer of the cervix—is very promising.

As of May 2003, 151 (79%) of 192 national members of the World Health Organization (WHO) had universal childhood vaccination programs. There are 89 member states that have been designated as having a high prevalence of HBV carriers. Sixty-four (72%) have universal infant vaccination programs. It is the goal of the WHO to have vaccination programs in all countries by 2007. The worldwide vaccination program is proceeding very well; it has saved millions of people from infection, illness, and death.

There are some unusual aspects of HBV infection related to sex differences. In general, males when infected with HBV are more likely to become carriers of the virus and females are more likely to develop anti-HBs (Table 3). Curiously, in areas of high HBV prevalence, the response of parents to infection with HBV is related to the sex of their offspring. In a study in Greece, families in which either parent was a carrier of HBV had a higher ratio of boys to girls than in families where the parents (particularly the mother) had anti-HBs (Table 4). Families with unaffected families had an intermediate ratio. Similar studies were done in five other populations; they were consistent with the initial results.


Total Number

Number Positive

Percent Positive
Marshall Islands, USTTPI
Cebu, Philippines
Manila, Philippines
Cashinahua, Peru

Data from Blumberg BS, Melartin L, Guinto RA, Werner B. Family studies of a human serum isoantigen system (Australia antigen). Am J Hum Genet 1966;18:594.


Live Births (Mean + SD)

Parent's Response to HBV
Couples (no.)
Sex Ratio
Either parent HBsAg(+): anti-HBs(−)3360 (1.8 ± 0.2)24 (0.7 ± 0.1)250 (161,429)*
Both parents HBsAg(−): anti-HBs(−)2951 (1.8 ± 0.2)35 (1.2 ± 0.2)146 (96,230)*
Both parents HBsAg(−): either parent anti-HBs(+)
24 (1.6 ± 0.1)
22 (1.4 ± 0.1)
109 (91,131)*

For definition of abbreviation, see Table 1.

From Blumberg BS. Sex differences in response to hepatitis B virus. Arthritis Rheum 1979;22:1261.

*Values shown in parentheses under “Sex Ratio” are the 5% confidence limits.

Emily Oster at Harvard University extended these family studies to populations (Figure 5). As predicted from the family studies, the ratio of males to females in a population is correlated with the prevalence of HBV carriers. In Alaska, there is a high prevalence of HBV carriers in the Native American Inuit population, but much lower levels in the American Indian population and the population of European origin. As noted above, there was a very successful vaccination program in Alaska. As predicted from the family studies, the male-to-female sex ratio decreased in the Inuit population, which experienced a drop in hepatitis B prevalence due to the vaccination program, but not in the other populations. This is consistent with the explanation that the HBV–sex relation is not only an association but that the HBV is the cause of the effect.

Changes in sex ratio have an effect on the demography, sociology, and economy of a society. These findings on HBV and sex ratio indicate that, in high HBV prevalence countries, such as China, the decreased number of females born, the so-called lost women demographic observation, may be due in large part to the high HBV prevalence and less so to other factors (Oster E. Hepatitis B and the case of the missing women. J Polit Econ [In press]).

We discovered HBV during the course of research on serum protein polymorphisms. Family genetic studies of the “Australia antigen”—that is, the carrier state for HBV identified by its surface antigen HbsAg—were done even before we were aware that we had discovered a virus. A summary of the family data analyses done on families in Cebu, the Philippines, and Bougainville, Solomon Islands, was consistent with the segregation of an autosomal recessive allele designated Au1. When present in double dose, it increased susceptibility to becoming a carrier of HBV (HBsAg positive) when the individual was exposed to the virus. Later, extensive investigations by Adrian Hill and others identified multiple human polymorphisms in which alleles segregating at the polymorphic locus increased the susceptibility to become a carrier of HBV. Many of these identifications were based on population studies. Alleles at the polymorphic locus, which were significantly more common in HBV carriers than in noncarriers, were inferred to increase susceptibility to the carrier state.

Often, alleles at the same susceptibility locus are related to response to several infections. These are arranged in Table 5 using published data available in the late 1990s. For example, individuals homozygous for the t allele at the vitamin D receptor locus (VDR) are more likely to become carriers of HBV and they also are more susceptible to pulmonary tuberculosis. Those who are homozygous for the alternate T allele, when exposed to the leprosy bacillus, are more likely to develop the lepromatous form, whereas those homozygous for t are more likely to develop the tuberculoid form of the disease. Another example is the tumor necrosis factor (TNF) locus. TNF is a cytokine (a protein or glycoprotein involved in the regulation of cellular proliferation and function) that has many roles, including the control of inflammation and the stimulation of the proliferation and destruction of cancer cells. There are several polymorphic sites on the TNF gene. They are related to susceptibility to HBV chronicity, cerebral malaria, lepromatous leprosy, and a form of the tropical disease caused by Leishmania braziliensis.






MHC class IIDRB1*1302HBV chronicity (P)Chrom 6Immune response
HBV response to interferon
Malaria, cerebral (P)
Papilloma virus
VDRttHBV chronicity (P)Chrom 12Vitamin D receptor binds vitamin D 1,25D3
ttTB pulmonary (P)
TTLeprosy, tuberculoid (P)
TtLeprosy, infection (P)
ttLeprosy, lepromatous (P)
ttBone mineral density, lower osteoporosis, increase risk prostatic cancer (P), inflamatory bowel disease
TNFG-308HBV chronicityChrom 6Tumor necrosis factor
Malaria, cerebral
Microcutaneous leish.
Lepromatous leprosy
MBPCodon 52HBV chronicityChrom 10Mannose binding protein
SLE, HIV (?) Infections in childhood
SM-1 5q31-33SchistosomiasisChrom 5Schistosomiasis susceptibility
? HBV chronicity
IL-10 promoterMutat. −108HBV chronicityChrom 1Immune response

Mutat. −059
HBV chronicity

Definition of abbreviations: Chrom = chromosome; HBV = hepatitis B virus; IL = interleukin; MHC = major histocompatibility complex; SLE = systemic lupus erythematosus.

There are interesting conceptual consequences of this grouping. These infectious agents are related to each other because they are related to the same susceptibility locus or loci. The infectious agents with affinities to the same locus constitute a microorganism gene affinity cluster. (I initially used the term “pathogen gene affinity cluster”; however, “microorganism” is more appropriate since some infectious agents are not ordinarily pathogenic.) The cluster provides another method of classifying microorganisms. They can vary from population to population since the susceptibility effects of the polymorphic alleles may be dependent on their environmental and genetic context. The infectious agents and diseases they may cause are likely to co-occur and the presence of one can provide clinical aids to seek out others in the cluster. In respect to the evolutionary effects of HBV, an understanding of the clusters gives insights into the complex interactions of specific infectious agents and their coevolution.

The alteration in sex ratio associated with (and probably due to) HBV infections is an example of a nonpathologic effect of the virus on its human host populations. Another example is the relation between HBV and iron storage. Patients with Down's syndrome who are HBV carriers have higher hemoglobin, hematocrit, and serum iron levels than those who are not (Table 6). In other populations, there is also increased serum iron levels in the carriers (Table 7). In populations with low iron intake in their diets, HBV carriers could be at an advantage compared with noncarriers in that they could retain more iron for metabolic use. This could be a positive selective factor, particularly in premodern times when populations had a shorter life expectancy and carriers would die before there were any clinical effects of the HBV infection.


Hemoglobin (g/100 ml)

Hematocrit (%)

Serum Iron (μg/100 ml)

TIBC (μg/100 ml)

Definition of abbreviation: TIBC = total iron binding capacity.

Increased hemoglobin, hematocrit, serum iron, and decreased TIBC in 20 patients with Down's syndrome with serum HBsAg compared with 20 without. The differences are significant.

Data from Sutnick AI, Blumberg BS, Lustbader ED. Elevated serum iron levels and persistent Australia antigen (HBsAg). Ann Int Med 1974;81:855.


Down's Syndrome

Renal Dialysis




Serum iron levels in (1) patients with Down's syndrome, (2) patients on renal dialysis, and (3) residents of a rural community in Senegal, West Africa.

Data from Blumberg BS. In: Szentivanyi A, Firedman H, editors. Viruses, immunity and immunodeficiency. New York: Plenum Press; 1986. p. 81–99.

The events and times for discovery and application are shown in Tables 8 and 9. The finding of Australia antigen in 1963 can be taken as the beginning of the research. By 1997, we had identified it as a hepatitis virus and confirmation from other laboratories followed quickly. (This was, in part, a result of our distributing the necessary reagents to those who requested them.) We invented diagnostic methods for identifying the virus in 1968, and subsequently (1975) were issued a patent for a radioactive method to greatly increase sensitivity. By 1969, donor blood was being tested for HBV and within the next few years, with the development of readily available commercial reagents, post-transfusion hepatitis due to HBV had largely been controlled in the United States and many other countries. We invented the vaccine in 1969 and the patent was issued in 1972. We soon (in 1975) licensed a nearby pharmaceutical company (Merck) to develop and produce the vaccine. The field trials for the vaccine were completed by Wolf Szmuness and his colleagues in New York City in the early 1980s, and the vaccine was approved rapidly (in 1982) by the U.S. FDA. National vaccination programs were in place by 1982; by 2003, vaccination was in use worldwide.


1957–1970s: Studies on serum protein polymorphisms and inherited susceptibility to disease
1963: Discovery of the “Australia antigen”
1967: Identification of HBV (Blumberg, London, Sutnick, Werner, Bayer, others)
1969 and following: Confirmation and extension of HBV identification (Okochi, Vierruci, Prince, others)
1968: Invention of diagnostic methods (Coller, Millman, Blumberg, Patent 3,872,225, issued 03/18/75)
1969: Invention of the vaccine for HBV (Blumberg, Millman, Patent 3,636,191, issued 01/18/72)
1969: Testing of donor blood to prevent post-transfusion hepatitis in Philadelphia and, later, elsewhere (Senior and Colleagues, 1974)
1970: Virus particle visualized (Dane, 1970)
1975: License for commercial production of HBV (Merck & Co, West Point, PA)
1980: Field trial of HBV vaccine (Szmuness and Colleagues, 1980)
1982: Approval of vaccine (U.S. FDA, and later, others)
1984: Start of national universal vaccination programs
1990: Decrease of primary cancer of the liver detected (Chang and Colleagues, 1997)
2003: Worldwide use of vaccine (WHO, 2004)

Definition of abbreviations: HBV = hepatitis B virus; WHO = World Health Organization.



Basic research to discovery of first evidence (“Australia antigen”)6
From first evidence to identification of HBV4
From identification of HBV to diagnostic methods1
From diagnostic methods to donor blood testing1 (+)
From identification of HBV to invention of vaccine2
From invention of vaccine to commercialization6
From commercialization to FDA vaccine approval7
From vaccine approval to national vaccination (Taiwan and elsewhere)2
From vaccine approval to worldwide use

For definition of abbreviation see Table 1.

There were 6 yr of basic research before we found the Australia antigen and an additional 4 yr before it was identified as HBV. The invention of the vaccine required an additional 2 yr. There were 10 yr between the issuing of the patent and its approval by the FDA; national vaccination programs began soon thereafter. Therefore, it required 27 yr from the beginning of the basic research to national programs and 36 yr from the basic research to worldwide use.

Similar but not exactly comparable dates for the polio vaccination program were supplied by Professor Francois Gros in the World Life Science Forum Newsletter for 2005. There was an interval of 70 yr from the basic science to the current status of near eradication of polio.

Hepatitis B vaccination is one of the largest worldwide disease-prevention programs. It has decreased the spread of HBV, particularly in China and East Asia. It has significantly decreased morbidity from liver disease and prevented the death of millions. HBV vaccination appears to prevent primary cancer of the liver; it is the first widely used preventive cancer vaccine.

There are important nonpathologic interactions of HBV with humans. Parents who are carriers of HBV have a higher ratio of males to females among their offspring than parents who developed antibody against the surface antigen. This may account for the high sex ratios seen in China and in other areas with a high prevalence of HBV infection. The apparent “loss” of females in these populations may be ascribed, at least in part, to HBV infection. HBV vaccination has decreased the ratio of males to females among newborns. If confirmed, this may have important biological, demographic, and economic effects. Another nonpathologic interaction is the relation of HBC to iron storage. HBV carriers have higher hemoglobin and hematocrit levels and appear to retain more iron from their diet than do noncarriers.

The response of the host to HBV infection is related to a series of polymorphic loci, which are, in turn, related to other disease-causing agents. Any changes in the prevalence of HBV may have an effect on the epidemiology of the other infectious agents.

Centers for Disease Control and Prevention, Atlanta, GA. Global progress toward universal childhood hepatitis B vaccination. MMWR Morb Mortal Wkly Rep 2003;52:868–870.
Centers for Disease Control and Prevention, Atlanta, GA. Acute hepatitis B among children and adolescents: United States, 1990–2002, JAMA 2004;292:2967–2968.
Blumberg BS, Friedlaender JS, Woodside A, Sutnick AI, London WT. Hepatitis and Australia antigen: autosomal recessive inheritance of susceptibility to infection in humans. Proc Natl Acad Sci USA 1969;62:1108–1115.
Blumberg BS. Australia antigen and the biology of hepatitis B. Science Natl 1977;197:17–25.
Blumberg BS, Gerstley BJS, Hungerford DA, London WT, Sutnick AI. A serum antigen (Australia antigen) in Down's syndrome leukemia and hepatitis. Ann Int Med 1967;66:924–931.
Blumberg BS. Hepatitis B: the hunt for a killer virus. Princeton, NJ: Princeton University Press; 2002.
Drew JS, London WT, Lustbader ED, Hesser JE, Blumberg BS. Hepatitis B virus and sex ratio of offspring. Science 1978;201:687–692.
Sun Z, Ming L, Zhu X, Lu J. Prevention and control of hepatitis B in China. J Med Virol 2002;67:447–450.
Correspondence and requests for reprints should be addressed to Baruch S. Blumberg, M.D., Ph.D., Fox Chase Cancer Center, 333 Cottman Avenue, Philadelphia, PA 19111. E-mail:


No related items