A Novel Microbe,Immunization Deaths,and Vaccination on Trial: BCG and the Lübeck Disaster of 1930

Abstract

Today’s controversies of gain-of-function virological research and mRNA COVID vaccination policies had an antecedent nearly a century ago in an event often referred to as “the Lübeck disaster.” From April through September 1930, 77 newborn infants in Lübeck, Germany, died after receiving oral BCG immunizations tainted with active human Mycobacterium tuberculosis. The tragedy threatened to end BCG immunizations. BCG and its originators, the French scientists Albert Calmette and Camille Guérin, were exonerated from liability. An inquest uncovered careless laboratory practices that contaminated doses of the vaccine with a human pathogen. The calamity underscored the necessity for exacting standards when handling dangerous infective microbes. The physician and bacteriologist responsible for the immunization program in Lübeck were tried and convicted for negligence, a concrete example of the civic responsibility expected of scientists when an experimental venture inflicts harm on an unwitting public. The example of Lübeck stands as an object lesson on the necessity of an informed, measured approach to any novel treatment. Controversies continue whether BCG vaccination is the preferred public health strategy against tuberculosis. Calmette and Guérin’s lasting scientific achievement is the creation of a microbe that over a century has kept its essential features of inciting a vigorous immunological reaction that was tolerated by its host and never regaining its pathogenicity. The features form the basis of modern cancer immunotherapy, where intravesical BCG is first-line therapy against non-muscle invasive bladder cancer.

Immunity Without Pathogenicity

When Robert Koch discovered the tuberculosis bacillus in 1882, he and the world’s best researchers raced to find a treatment for the dreaded disease. Among them was Albert Calmette (1863-1933), a former pupil of Louis Pasteur and director of the Institut Pasteur laboratories in Lille, a city near the French border with Belgium. Working with him was Camille Guérin (1872-1961), the head of the laboratories at the institute.

Calmette and Guérin drew inspiration from Edward Jenner’s work where he used a nonhuman pathogen, cowpox, as a vaccine against smallpox (1796). They decided to use Mycobacterium bovis, isolated from tubercular cattle, as the basis for a vaccine against human tuberculosis.

Beginning in 1908, Calmette and Guérin subjected their sample of M. bovis to successive culture passages and chemical stresses with the aim of retaining its immunological properties while neutralizing its pathogenic effects. Their strategy recapitulated Pasteur’s methods from earlier vaccine developments against chicken cholera (1879), ovine anthrax (1881), and rabies (1885).^1,2

By 1913, they developed a strain that conferred immunity to mycobacterial infection without causing disease. Calmette named the microbe “Bacillus Calmette-Guérin,” or “BCG.” Through years of experimentation, Calmette and Guérin confirmed the efficacy of BCG across various animal models. Most important, the organism never regained its prior virulence even after extensive culture exchanges and inoculations. ¹

“The Last Verse Sung by the Mother at the Cradle”

In a 1926 monograph, Calmette documented their research and early clinical experience. ¹ Calmette wanted to immunize newborn infants, especially those born to mothers with active tuberculosis. Under such circumstances infection with tuberculosis was nearly guaranteed, with 76 percent turning their tuberculin skin test positive within the first year of life and a frighteningly high mortality of 25 percent, a figure based on data from Belgium and England. Calmette quoted Emil von Behring on the poignant circumstance by which infants contracted tuberculosis from their mothers: “It is almost always the last verse of a romance sung by the mother or nurse at the cradle.” ¹

Convinced of the efficacy of BCG immunization, Calmette was thus ready to use it in humans despite ongoing animal experiments. In July 1921, a request from Benjamin Weill-Hallé, a physician at the Hôpital de la Charité in Paris, provided an opportunity to administer BCG to a baby cared for by a grandmother who had active tuberculosis. A tiny 2 mg oral dose was given, followed by 2 additional doses given 2 and 4 days later that completed the course. Remarkably, the baby had no ill effects and never contracted tuberculosis.

Buoyed by this success, Weill-Hallé immunized more infants with a higher dose in 1922. Ultimately, 217 infants were given BCG with no reported fatalities from tuberculosis. By early 1926, more than 5000 newborns across France had been vaccinated.

The extensive acceptance of BCG complicated any formal comparisons with unvaccinated children. As a backdoor proof of its efficacy, Calmette distributed questionnaires to practitioners who used the vaccine, asking whether their patients contracted the disease and if the child died, the cause of death.

From his responses he determined that only 0.7 percent of children who received BCG died of tuberculosis; among those who stayed in households where someone had active disease, the mortality was a mere 1.8 percent. Compared to his estimated mortality of 25 percent, the efficacy of BCG vaccination in Calmette’s mind was proved. ¹

Calmette’s reputation soared as BCG gained favor with physicians in France and Scandinavia. By 1930, an estimated 400,000 infants, including nearly a quarter million in France alone, had received BCG without incident. ³ Many considered his work was worthy of the Nobel Prize. ⁴

Opposition

BCG, however, faced opposition. In 1927, the German Government Health Advisory Council determined that the evidence was insufficient to endorse BCG. ⁵ A panel of expert statisticians convened by the Health Organization of the League of Nations (the forerunner to today’s World Health Organization) called for formal prospective clinical trials with control groups, requirements that health agencies worldwide had begun to mandate for new treatments. ⁴

In 1928, writing in the British Medical Journal, Major Greenwood, professor of epidemiology at the London School of Hygiene and Tropical Medicine, asserted that the mortality of infants born to tubercular mothers was closer to 7.7 percent, with the overall mortality among infants below 1.8 percent. ⁶ He condemned the methodological flaws in Calmette’s data as lacking scientific rigor. “It cannot be regarded as a serious contribution to scientific literature at all,” Greenwood wrote. ⁶ He paraphrased George Bernard Shaw’s quotation that “Silence is the most perfect expression of scorn:” ⁷

Dr. Calmette’s excursion into statistics might properly have been received with that silence which falls upon ordinarily good-natured people when a man of distinction offers, as his contribution to the discussion of a matter he has never studied, an absurd blunder. ⁶

Moreover, in 1929 Strashimir Petroff of the Trudeau laboratories in Saranac Lake, NY, alleged the emergence of a virulent BCG subtype under experimental conditions, ⁸ although his claim was later disregarded. To many it confirmed the nightmare scenario of the injection of pathogenic organisms into unsuspecting innocents.

That same year A. Stanley Griffith, a bacteriologist to the Medical Research Council (MRC) of Great Britain, advised against the use of BCG in livestock, much less humans. The MRC thus rejected BCG for vaccination. A Canadian remarked in 1932 that the initials “BCG” could stand for “Better Go Cautiously.” ⁹

Unfolding Tragedy

Health officials in Lúbeck, a city on the Baltic coast in Northern Germany, failed to heed the warnings in what Peter Donald of the University of Stellenbosch, South Africa, called an “unfolding tragedy.” Donald’s article on the Lübeck disaster is the source for the facts in the following section. ¹⁰

In 1929 George Deycke, director of the Lübeck General Hospital, sought to implement BCG immunization in his tuberculosis dispensaries, assisted by bacteriologist Ernst Alstädt. In July Calmette sent BCG to Lübeck, Alstädt placing them within the hospital laboratories.

The facility also housed active tuberculosis cultures. Deycke and Alstädt had obtained samples of the “Kiel strain” from bacteriologists at the Charité Hospital in Berlin years prior for their experiments on M. tuberculosis proteins. The Kiel microbe had undergone so many passages that the experts from the Charité believed it had lost most of its virulence. The strain also acquired the unique property of staining Sauton growth medium green, a characteristic that later identified it as a contaminant in BCG cultures in Lübeck.

Unaware of the time bomb in his hospital laboratory, Deycke urged the town council to start an immunization program using BCG. Among from the hundreds of thousands that had ingested the microbe without incident were several hundred infants in Germany.

In February 1930, the Lübeck town council and the local practitioners’ organization approved Deycke’s plan to inoculate the town’s newborn infants. Advertisements promoted the program. Three infants were inoculated in December and January ahead of the planned launch date in late February.

A warning sign emerged in mid-January 1930 when one of the early inoculated infants fell ill with cervical lymphadenitis, later confirmed by biopsy to be infected with human M. tuberculosis. Ignoring the evidence, Deycke assumed that the baby had a congenital infection, and the BCG campaign went on as planned.

On April 17 the first vaccinated infant died, but the parents refused an autopsy. When another baby died 3 days later, authorities insisted on a postmortem examination. The baby had disseminated disease, with extensive respiratory tuberculosis and mesenteric nodal infiltration. Doubting the possibility of an intestinal route of infection because of the disease in the lungs, Deycke concluded that child must have contracted the disease via the respiratory route since the mother had active pulmonary tuberculosis.

Then, 2 more infants died on April 25 and 26. When both were found with severe abdominal tuberculosis, the only plausible explanation was the ingestion of active tuberculosis organisms from the purported vaccine.

Following the death on April 26, the Lübeck laboratory halted its BCG program. Deycke also destroyed his stocks of BCG. The incriminating act that hindered later investigations, but enough evidence was salvaged for scrutiny. ¹¹

In an egregious oversight, distributed vaccines were not recalled, allowing infants to continue receiving second and third doses of the contaminated material for 4 crucial days until the inoculations were completely stopped on April 30. ¹²

The few deaths in April more than tripled in May, with numbers continuing to rise throughout the summer. Death struck with alarming speed, occurring at a median of only 42 days after inoculation. By September 30, the outbreak had largely subsided.

Of 412 babies born in the city in 1930, 251 received the vaccine. Ninety percent (228) showed clinical or radiological evidence of human tuberculosis and 30 percent (77) died. Of the 174 survivors (69%), 44 (25%) escaped clinical signs or radiological evidence of disease. Most of them (126, 72%) had calcifications visible on abdominal plain films, indicating a gastrointestinal route of infection. Six (3.5%) were left with permanent hearing loss as a sequela of their infection. ¹⁰

Inquest and Trial

Lübeck officials finally informed the Health Ministry on May 14. On May 21, the federal minister of the interior committed a team to thoroughly investigate the outbreak, from the source of the microbe at the Institut Pasteur to every infant that received BCG. ¹¹

The probe, led by Albert Moegling of the German Tuberculosis Research Institute, included representatives from the Federal Ministry of the Interior, the Federal Health Bureau, and the Robert Koch Institute. Independent physicians from Cologne examined the surviving infants and the postmortem results of those that died. ¹⁰

The infants were infected with the Kiel strain of M. tuberculosis, incriminated by the green stain in the culture medium. BCG intended for inoculations had somehow become contaminated with pathogenic M. tuberculosis. ¹³

In his 1926 monograph on his BCG research, Calmette outlined his exacting instructions on the care of BCG cultures: keeping cultures in a separate incubator exclusively for BCG samples, unique identification of BCG cultures using labels of its own color, and animal testing to verify that it had not regained its pathogenicity. ¹

The Lübeck lab followed none of the procedures. No laboratory notes were kept, making impossible to trace how the samples got mixed. BCG and virulent tubercle bacilli were handled in the same room by the same person, who was not a trained laboratory technician.^3,11 Sloppy laboratory methods, wrote Henry Willis in the American Review of Tuberculosis, “allowed such a mistake to creep into reality.” ¹⁴

From the investigation, prosecutors charged Deycke, Alstädt, Max Klotz, chair of the Lübeck health department, and Anna Schütze, Deycke’s laboratory assistant, with criminal negligence. The trial began in October 1931 and attracted so much attention that court was held in a school gymnasium.

It was an international story covered by both professional journals and the lay press. In the words of a correspondent of the Journal of the American Medical Association (JAMA), Lübeck “was one of the most tragic events in the history of medicine … a case in which the whole civilized world took an interest, and which, on several occasions, caused great excitement among the general public.” ¹³

A verdict was reached on February 6, 1932. Deycke was sentenced for 2 years in prison for homicide due to negligence; Altstäedt, 1 year and 3 months for bodily injury due to negligence. Klotz and Schütze were acquitted. ¹³

The JAMA writer had sympathy for Deycke and Alstädt. They were diligent in their work against tuberculosis on behalf of the city, he wrote. “[They] were certainly actuated by the best motives. … [yet at trial] account was taken of the extraordinary nature of the disaster, the death of so many children, the sufferings of those taken ill, and the grief of the parents.” ¹³

Aftermath

Calmette

Calmette’s reputation was harmed by the scandal. An attorney for the plaintiffs denounced Calmette as the person chiefly responsible for the outbreak. The prosecutor chose not to try the French scientist because his published protocols had not been followed and so many throughout the world had already received BCG without incident. ¹³ Calmette did not testify at the trial but defended himself in the German medical press and at international professional meetings. ¹⁵

Nettled by calls for randomized controlled trials by people he considered his inferiors, he initiated a controlled trial in the French colony of Algeria where tuberculosis was rampant. Conducted from 1930 to 1956 and ended because of the Algerian War (1954-1962), the Algerian BCG trial was the largest study of its kind. ⁴ The results, made public in 1960, were inconclusive. ¹⁶

Calmette never saw the study to its completion. Drained by the criticism of his work and his defense of his role in the Lübeck disaster, he died in October 1933 after a brief illness. ⁵

“Calmette Kinder”

The survivors, called the “Calmette Kinder,” received an outpouring of kindness. ¹² Many suffered active disease until the age of 3 to 4 years. The Lübeck government made sure the victims were well nourished and sheltered, giving their families food supplements and rent subsidies. As compensation the children were awarded savings books with sums of money assigned according to the severity of illness, from which they could make withdrawals in 1951 when they reached age 21 years. With the fall of the Reich in 1945, Reichsmarks were converted to Deutschemarks at a 10-to-1 ratio, making the modest nest eggs nearly worthless.

Especially affected were those made deaf from infection of the middle ear. One girl suffered facial nerve paralysis in addition to her hearing loss. Stigmatized by her appearance and unable to hear or speak, she attended a specialized school for the deaf and received a pension for life. She had a permanent job in the sewing room at the Lübeck General Hospital. ¹²

BCG as a Vaccine Against Tuberculosis

All BCG vaccinations in Germany halted because of the events at Lübeck. “The rejection of BCG in Germany can be seen as a rejection not just of BCG in Germany but of French science,” wrote Linda Bryder, historian at the University of Aukland. She quoted Francis Barrymore Smith, historian at the Australian National University: “[The Lúbeck disaster] was a fitting backdrop to the rise of the Nazis to power in Germany in the early 1930s.” ⁹

Amid the unchecked spread of tuberculosis that followed World War II, the Danish Red Cross led a BCG vaccination campaign throughout the European theater from 1945 to 1948. Germany chose not to join. ¹⁷ UNICEF and the WHO saw tuberculosis as a health issue affecting all countries.

In 1948 they initiated the International Tuberculosis Campaign, a worldwide rollout of BCG vaccinations. ¹⁷ The results of the immunization campaigns were wildly divergent, with estimates of prevention ranging from 0 to 80 percent. ¹⁸

One problem was that Calmette and Guérin’s original organism was never cloned. It had been cultured and re-cultured under different conditions in multiple laboratories. Strains evolved with different phenotypic and genomic features that may have accounted for the differences in efficacy. Today several vaccine companies manufacture BCG for worldwide consumption. “There is insufficient data,” wrote Simona Luca of the University of Medicine and Pharmacy in Iasi, Romania, “to favor or recommend one BCG vaccine strain.” ¹⁵

A systematic review of randomized controlled trials of BCG by Punam Mangtani and associates at the London School of Hygiene and Tropical Medicine in 2014 found that BCG protection among children with non-reactive tuberculin skin testing had a rate ratio (RR) of 0.26 and a 95% confidence interval (CI) of .18-.37; among infants the RR was 0.41 and CI: .29-.58. ¹⁹ Today BCG is still the only vaccine against tuberculosis and is listed by the World Health Organization as one of its essential medications. ²⁰

BCG as Immunotherapy Against Bladder Cancer

When instilled in the bladder BCG stimulates cellular immune mechanisms that kill mucosal neoplasms in the bladder, a therapy first reported by Alvaro Morales, of Queen’s University in Kingston, Ontario, in 1976. ²¹ Since then, a myriad of microbial products and monoclonal antibodies have been tested as potential agents against bladder cancer. Nearly 50 years after it was first used, intravesical BCG remains the standard intervention for intermediate and high-risk non-muscle invasive bladder cancers.

An Object Lesson

The Lübeck disaster is a warning on the hazard inherent in conducting research on pathogenic microbes. Caution and informed consent are absolutely required when implementing novel treatments on the public. Calmette and Guérin, as the microbe’s creators, and public health officials like Deycke, who implemented wide-scale vaccination programs, had the ongoing responsibility for the well-being of the millions who were given the microbe as a vaccine.

Calmette and Guérin’s scientific achievement was the creation of a bacillus that induced a durable immunological response without causing disease. Modern clinicians took advantage of this capability and used it against bladder cancer. In contrast to the disputes regarding its original use as a vaccine, the role of BCG in cancer immunotherapy is unquestioned.