There are two principal types of diabetes which are very different diseases, though they have similar symptoms. Type 1 diabetes is an autoimmune disease where the body's ability to make insulin is eliminated as insulin-producing cells are destroyed by the body's own immune system. Type 2 diabetes is quite different and results because the body's ability to respond to insulin is damaged in a variety of ways.

Newspapers are currently reporting that use of the BCG vaccine (a formerly common childhood vaccine that protected, among other things, against tuberculosis) can cure type 1 diabetes. For example, The Daily Telegraph reports:

The BCG vaccine against tuberculosis can reverse Type 1 diabetes to almost undetectable levels, an eight-year study has shown.

US researchers found that just one jab, followed by a booster four weeks later, brought down average blood sugar levels to near normal within three years, and the effect lasted for the following five years.

It is hard to tell from newspaper reports how credible this study is. Given what we know about the nature of diabetes are the reports credible? Could BCG be a cure for type 1 diabetes?

  • 2
    I've not got time right now to write an answer - but if someone else feels like it here's the report: nature.com/articles/s41541-018-0062-8
    – Jamiec
    Commented Jun 22, 2018 at 10:29
  • 3
    It is worth noting that the trial doesn't, strictly interpreted, report a cure for diabetes which would require a restoration of insulin production in the pancreas. What it actually reports is a very significant improvement in the average blood sugar levels and a large reduction in side effects like hypoglycaemic events in diabetics still taking insulin.
    – matt_black
    Commented Jun 22, 2018 at 10:58
  • 2
    My wife is Type 1 and follows such stuff fairly closely (as do thousands of others in the US). There has been some intriguing research in the past 5 years, but nothing has panned out to date. Commented Jun 22, 2018 at 12:00
  • 1
    Downvotes already! I'll improve the question if you tell me why.
    – matt_black
    Commented Jun 22, 2018 at 12:10
  • 3
    @Fizz Theory of diabetes isn't very relevant to the question. But I think what I said was broadly correct: the defining characteristic of type 1 is autoimmune destruction of the body's ability to produce insulin; type 2 diabetics still produce insulin but either not enough or their body doesn't use if properly. The diagnostic distinction is whether insulin is still produced by the pancreas.
    – matt_black
    Commented Jun 22, 2018 at 19:07

3 Answers 3


I work in Diabetes research in the UK and Can find no evidence to support the wild claims in the paper, press release and press reports. Sadly hidden away in a table in supplementary data is the fact that the result of improved glucose is a result folrom 3 people who had the vaccine compared to 3 had placebo. This very small number makes it very likely that the result is just chance and not a real improvement. This is supported as the mechanism is implausible as there is no evidence of any retained insulin production.

Here is the paper to show the sample size. https://www.nature.com/articles/s41541-018-0062-8#Sec8 I don’t think any references are needed to show that very small studies 3 cases and 3 controls have a very high level of both Type 1 and Type 2 errors

If a study is prospective does not alter its power so if the sample size is tiny then the power is very low.

The authors claim that their results are statistically significant- this is unbelievable as the difference between the vaccine and the non vaccine groups HbA1c are only 0.8%different 6.4% v 7.2% As the SD in HbA1c is typically 1% this is a difference of around 0.8SD.

  • 6
    What you say may be correct, but it needs to be referenced to be a good answer here. Link to the report (and page) for the values and possibly link to normal standards of experimental design to show how dodgy very small trials are.
    – matt_black
    Commented Jun 22, 2018 at 14:23
  • Welcome to Skeptics! Please provide some references to support your claims.
    – Oddthinking
    Commented Jun 22, 2018 at 18:48
  • 1
    @AndrewH: I am keen to have this answer expanded; we love hearing from people with direct expertise, but we can't rely on an appeal to authority, so if you can point out how you reach your conclusions, it would be great. The n=3 sample sizes are mentioned in the paper itself they claim to have statistically significance. They say it is a prospective study, which is probably a key part of the answer.
    – Oddthinking
    Commented Jun 23, 2018 at 3:52
  • 3
    That the supported mechanism is implausible is currently your personal opinion. You may well be right, and you certainly understand this better than I do, but we are looking for sources we can check for ourselves.
    – Oddthinking
    Commented Jun 23, 2018 at 3:54
  • 2
    Thank you for making the edits. I appreciate you try to improve the answer. The edits are still unconvincing though. If someone made an n=3 study that showed vaccines would make amputated limbs grow back, it would be huge news. So, the fact that n=3 itself isn't enough. You claim the SD of HbA1c is typically 1%. That needs a reference. Also, a reference to what the heck HbA1c is and why it is relevant. Your claim that the mechanism is implausible needs a reference.
    – Oddthinking
    Commented Jun 24, 2018 at 2:46

No. The study doesn't claim that and the results are very weak anyway

This story is an interesting case where the headlines exaggerated the claims and the claims were far weaker than many supposed anyway.

What the study actually claimed

The study didn't claim that anyone had been cured from type 1 diabetes. The paper says:

This 8-year-long Phase 1 trial shows the long-term lowering of blood sugars after year 3 measured by HbA1c. The lowering of blood sugars to a range near normal was maintained for the next 5 years of monitoring... ...Taken together, the findings suggest the benefit of targeted reintroduction with attenuated Mycobacterium bovis as a BCG vaccine for blood sugar control through epigenetics and altered metabolism.

The claim is that some type 1 diabetics (type 1 diabetics have lost the ability to produce their own insulin) achieved better blood sugar control:

BCG treatment subjects with the improved and tight blood sugar control demonstrated blood sugar stability and lowered blood sugar persistently for 5 continuous years after the initial drop in values. Semi-annual surveys confirmed that during year 03 to year 08 after BCG vaccinations there were no reports of severe hypoglycemia by any patient, even with lowered HbA1Cs near the normal range, and no change in their care as it related to new insulin pumps or continuous glucose monitoring devices. The placebo group of subjects continued to show hypoglycemia events during the same time periods of monitoring.

...In the human, this stable blood sugar control was not driven primarily in these human subjects by pancreas recovery or regeneration.

The subjects continued to take insulin so there is not claim that their diabetes was cured. Their blood-sugar control is claimed to be significantly improved (both the average blood sugar was lowered and the number of incidents caused by extreme low sugar were reduced (these normally are more common in diabetics with lower blood sugar). Both these results would be significant improvements for type 1 diabetics as those who achieve lower levels of blood glucose are at higher risk of debilitating extreme low sugar events.

So the paper is claiming significant improvements in sugar control but not a cure.

But the study is statistically weak, involving a very small number of patients

The headline summary of the study (which investigated the mechanism of actions as well as the major effect on blood sugar control) claims a reasonable number of subjects for an initial study:

This is a study of 282 human research participants for both in vivo BCG vaccine clinical trial studies (n = 52) and in vitro mechanistic studies (n = 230). Of these total research subjects 211 had type 1 diabetes and 71 were non-diabetic control subjects.

This sounds good but a close reading suggests that many of these subjects were not enrolled in the key part of the study but were used to investigate the potential mechanism of action ( the study proposes a new glucose-consuming mechanism is activated in blood cells by the vaccine). The number where the outcomes were observed was much smaller:

For all BCG treated subjects followed for up to 5 years, in total of 52 subjects, their clinical traits are represented in Supplementary Table S1a. Some BCG and placebo treated Type 1 diabetic subjects were enrolled and were studied for 8 years (Supplemental Table S1b, n = 6 subjects); other BCG treated (n = 9), placebo treated T1D (n = 3) and reference T1D (n = 40) were studied for up to 5 years...

So 52 subjects were studied for 5 years but only 6 for 8 years. Further reading of the paper shows that the key results of persistently reduced blood sugar come from a comparison of just three patients who are compared to just three controls.

Moreover the gains reported, while significant, are far from a breakthrough. The reported long term blood sugar levels are better (~10% better than the control group) but would need to be 30-50% better to reach the normal levels in non-diabetics. And while the paper does not report this explicitly, the confidence that the result is caused by the BCG vaccine and not by chance have to be low because the study is so small. It may be the study will report a similar result in another three years when longer follow ups have been concluded at which time more confidence might be justified. But three patients is not a large enough group to claim a breakthrough now.

Conclusion: certainly not a cure and possibly not a breakthrough we can be confident of

The study claims a modest improvement in blood-sugar control. But in such a small number of patients the it is hard to be confident the result in not a statistical fluke. Moreover, the gains are not a breakthrough, merely a modest improvement. The headlines are not justified.

A summary of the research by the charity Diabetes UK mirrors these conclusions.

  • This kind of thing, repeated, is soul-destroying for diabetics and their families.
    – Benjol
    Commented Sep 11, 2018 at 7:38

Andrew H's answer is only somewhat correct, the 3 patients were only the first phase of the first phase (ha), the grand total seems to be 9; still this treatment is controversial, something that was hinted well enough in the source used for the question (at the end of the article, several experts cautioned about the small sample size). Another article employes thinner gloves... on both sides of the issue:

While the results, from only nine patients, must be replicated in a larger study, said Dr. Joseph Bellanti of Georgetown University Medical Center, “if what they found is true, they really have something here.” Bellanti, who was not involved in the research, said the study’s eight-year follow-up and use of a placebo control arm made him “cautiously optimistic” that two doses of the BCG vaccine “can decrease levels of A1c,” a measure of blood glucose that predicts the likelihood of serious complications such as stroke and kidney failure.

JDRF (formerly the Juvenile Diabetes Research Foundation), the Joslin Diabetes Center, and several university diabetes centers all declined to speak about Faustman’s results. She has been a voice in the diabetes wilderness for nearly two decades, angering the establishment diabetes community by pursuing low-tech research very different from more popular approaches, such as embryonic stem cells and immunosupression.

Critics have gone so far as to send letters to newspapers that covered her work apologizing to patients “on behalf of Dr. Faustman” for “having their expectations cruelly raised.” She has also struggled for funding, receiving much of her research support from the private Iacocca Family Foundation, rather than in federal grants.

In Faustman’s Phase 1 clinical trial, three participants with type 1 diabetes received two doses of BCG vaccine, a month apart. After the vaccine showed signs of effectiveness, an additional six patients were vaccinated five years ago, and 111 more recently. The new paper and a presentation scheduled for a meeting of the American Diabetes Association this weekend focus on the patients who have been followed for more than five years.

“We wanted it to be good, but we didn’t know it would be this good,” Faustman said.

All of the patients remain on insulin, she said, but less of it. They are also able to monitor their blood sugar less frequently, which can be several times an hour. (The standard of care is a continuous glucose monitor, in which a probe is inserted into the abdomen, plus an insulin pump.) “If we can gradually move people to where they can control their blood sugar, their minute-to-minute lifestyle can improve dramatically,” Faustman said.

The A1c reductions could also bring significant health benefits. Every 10 percent drop, research shows, reduces complications such as stroke and heart attack by about one-third. The BCG vaccine lowered A1c levels 9 percent to 16 percent.

A Phase 2 clinical trial of BCG is currently underway at Mass. General. It is testing multiple BCG doses in 150 patients with longstanding type 1 diabetes.

Although many studies of BCG are underway around the world in diabetes and other autoimmune diseases, including multiple sclerosis, there has been little interest among U.S. researchers outside Faustman’s lab.

“There is not a lot of enthusiasm because we’re all rewarded for discovering for-profit drugs,” she said. “Potential funders come [to my lab] and ask, ‘How can we make money off this?’”

BCG, whose one licensed manufacturer in the U.S. is Merck subsidiary Organon Teknika, costs less than a dollar a dose. (Faustman used a strain made by Sanofi.) The U.S. market for insulin meters and insulin pumps is $20 billion. “With everyone thinking they need a pump and a meter, if you come along with an inexpensive vaccine that can change this standard of care, of course there will be pushback,” Faustman said.

So I doubt this will have a satisfactory answer until the larger (150 patient trial) concludes.

Also, Reuters reported some putative mechanisms of action being investigated and perhaps even replicated (by others) in mice, so Andrew's part of the answer on that seems incorrect. I'm not in a position to evaluate the plausibility of the mechanism myself, and certainly some other researchers appear to have dissed it:

The study is nevertheless within the mainstream of current efforts to cure diabetes, said cellular immunologist Raphael Clynes of the Berrie Center, who was not involved in the research.

"Folks had presumed that by the time patients had overt diabetes, all their islet cells had been destroyed," Clynes said. "We now know there are preserved islet cells many years out. The presumption is, if you can get rid of the inflammatory, autoimmune response, the islet cells could regenerate."

Faustman's research on lab mice, begun in the 1990s, suggested that one way to get them to do that was with a compound in the body called tumor necrosis factor. TNF is difficult to obtain commercially, but the BCG vaccine increases the body's production of it.

Raising levels of TNF through the vaccine cured type 1 diabetes in mice, Faustman reported a decade ago: With the autoimmune T cells out of the picture, islet cells regenerated and began producing healthy amounts of insulin. The journal that published the study insisted Faustman not use the word "regeneration," so controversial was the idea that islets could come back to life. She substituted "restoration."

By 2006, other scientists - including some who had attacked Faustman's claims of regeneration - had replicated key findings of her mouse studies, laying the foundation for the clinical trial.

Three patients with longstanding diabetes received two injections, four weeks apart, of BCG. Three others received saline injections, serving as the placebo control.

In two of the three BCG patients, levels of islet-attacking T cells fell, the scientists reported on Wednesday. Dead autoimmune cells were released into the bloodstream, a hint that TNF was killing them as intended. A measure of insulin production rose. The only placebo patient with similar results had become infected with the Epstein-Barr virus, which also triggers production of TNF.

Althoug in other respects, the Reuters coverage reiterates the criticism with a bit more detail:

Faustman has already faced significant challenges to her theory. JDRF rejected her funding requests and circulated a 2003 letter from two of her colleagues at Harvard Medical School, casting doubt on her work and apologizing to diabetics for "having their expectations cruelly raised" by stories about her research.

Reaction to this study was not much better. "The paper shows that BCG is associated with a transient improvement in a couple of patients, but it's hard to conclude that TNF is the causative factor," said Columbia's Clynes.

"It's certainly interesting and worth further investigation," said JDRF Chief Executive Jeffrey Brewer. "But it's really important to be careful about how we interpret early results." Curing type 1 diabetes may finally be within reach, he said, "but it will be a marathon, not a sprint.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .