Regenerating in Hannover, Part 2: Axel Haverich’s “growing” heart valves

The science of the fallen star of regenerative medicine Paulo Macchiarini was simple: take a dead organ, strip it of its cells and seed the carcass with stem cells (usually the magic cells from bone marrow). After some days in a “bioreactor”, you take out a living trachea, esophagus, even heart, and implant it into a patient. Another human life saved, and not only media, even scientist colleagues fell for this outrageous quackery. As the result of this hubris, several patients died, others remained in permanent critical care. Macchiarini and his partners Philipp Jungebluth, Martin Birchall and others had to fake ethics approvals as well as to lie and cheat about medical records in their publications in The Lancet, all in order to present a miserably suffering recipient of a ”regenerated” trachea as fully recovered. Animal experiments were performed only after Macchiarini’s team operated their first human patient, as indirectly evidenced by Jungebluth’s own doctorate thesis at the Medical University Hannover (MHH) in Germany.

Macchiarini began to develop his “decell-recell” method of organ regeneration while working in Hannover, close to the renowned heart surgeon and MHH clinic director, Axel Haverich (see Part 1 for the background). In 2009, the Italian cheater then moved on to a professorship at the Karolinska Institute in Stockholm, where he was showered in funding money and received best institutional protection, despite his patient abuse and his lies about his qualifications.

Big dreams

Both Macchiarini and Haverich were obsessed with the idea of growing a living beating heart using stem cells from a dead decellurised donor organ (see Haverich in Circulation, 2009). This bombastic dream Haverich eventually had to abandon, but his more modest one, namely that of regenerated growing heart valves, seems to be a full success. German head of state nominated Haverich and his surgeon Serghei Cebotari in 2008 for the Zukunftspreis (Future Prize), based on the recommendation of the central public funding agency Deutsche Forschungsgemeinschaft (DFG). Also the European Union was most impressed: Haverich, his private company Corlife and the research consortium ESPOIRE under his leadership received in 2011 €6.6 Million (of this €5.5 Million directly from the EU). The clinical trial started in January 2012 and will conclude by the end of this year. The basis for ESPOIRE was Haverich’s own clinical research in Germany and in the former Soviet republic of Moldova, where a certain Anatol Ciubotaru (father of Haverich’s co-author Serghei Cebotari) heads a major clinic in Chisinau. In this regard, literature evidence suggests human experiments in Moldova were performed prior to animal testing in Germany.

This is how the EU describes their funding rationale for ESPOIRE:

“Acquired and congenital heart disease can necessitate heart valve replacement. However, current heart valve substitutes are not considered ideal as they need anticoagulation, bearing the risk of bleeding when manufactured from non-organic material, or they degenerate when they derive from animals or human tissue donators (homografts) thereby leading to frequent reoperation especially in the young population. An ideal heart valve substitute would overcome these limitations and even have the potential to grow when implanted in pediatric patients.

Haverich et al. have developed an implant for heart valves, which is better tolerated than the known alternatives and which has the potential for regeneration by autologous recellularization. Implants derive from donated homografts, which are chemically treated to inactivate potential microorganisms and viruses. The heart valves then are decellularized chemically, so that only connective tissue remains, the matrix of the decellularized heart valve.”

Dead tissues come alive

Key words here which make Haverich’s heart valves so special (compared to anything else on the market) are “autologous recellularization” and “potential to grow when implanted in pediatric patients”. The first one means, these dead pieces of connective tissue are somehow brought back to live inside the patient’ hearts by their own blood and its circulating endothelial stem cells: a miracle of human blood’s regenerative capacity which the rest of biomedical scientists had been unaware of for centuries. Such miraculous revival of dead heart valves is expected to allow them to grow together with the young patients. No need to regularly replace the worn-out or undersized valve transplant (as it is currently an unfortunate necessity), since Haverich’s valves are supposed to become a living growing part of the patient.  Yet, as my discussion with several cardiovascular experts and a close study of Haverich’s own publications showed, neither of these two claims were actually ever proven by him scientifically. Despite this, the ESPOIRE website announced that Haverich’s miracle valves will be tested

“in direct comparison to current heart valve substitutes, such as cryopreserved homografts and xenografts, within a large prospective multicentre trial at 8 leading European Centres for Congenital Cardiothoracic Surgery At least 200 patients are needed for robust statistical analysis regarding re-operation and re-intervention rates, hemodynamic performance, growth potential and long term durability”.

The valves are being manufactured by the Hannover-based private company Corlife, which was founded in 2006 by Haverich “as a spin-off from the MHH” and is still led by him, together with the managing director Michael Harder. But what exactly is the science behind these regenerated heart valves?  Haverich himself summed up his achievement in April 2013 in a PowerPoint presentation titled “Our long journey to ESPOIR”, which I found on the MHH website (copy here).

Humans first, animals second?

Just like Macchiarini, Haverich apparently first tested his regenerative technology in humans before moving to animal models such as sheep. He did not choose German patients though, maybe because of the tough ethics and medicinal product control. Haverich and his surgeon Cebotari Junior went to the Cardiac Surgery Centre at the State University of Medicine and Pharmaceutics in the Moldovan capital Chisinau, run by Cebotari Senior. Moldova is a desperately poor and corrupt country, riven by a bizarre pro-Soviet cessation conflict since 1990 and recently ranked as a sad 103^rd in a list 168 states by Transparency International. It is somewhat telling that its citizens look up to Romania (of which Moldova used to be part of before Soviet annexation) as a paradise of economical possibilities. Yet, Cebotari et al proudly suggested in their study in 2006 Circulation that a Moldovan ethics approval was just as good as elsewhere:

“The study was performed in accordance to the Law on Health Care System (No. 411-XIII from 28.03.1995) of the Republic of Moldova. The Council of Experts of the Ministry of Health (No. 09 to 1/2 from 02.05.2002) as well as Ethical Medical Committee of the Ministry of Health (No. 01 from 17.05.2002) of the Republic of Moldova approved the study”.

In this human experiment from 2002, two Moldavian children aged 11 and 13 received decellularized pulmonary valves, which Haverichs team claimed to have regenerated with the “autologous endothelial progenitor cells” present in the blood. The word “autologous” normally suggests that the blood was each patient’s own, however, the Materials and Methods section of the paper speaks of “blood of healthy human volunteers”, thus possibly not that of the patients after all. Which does make sense logistically: the valve regeneration was performed not in Moldova, but in Haverich’s Hannover lab at LEBAO (Leibniz Research Laboratory for Biotechnology and Artificial Organs). There, the decellurised cadaver valves were placed inside a bioreactor and washed with culture medium spiked with cells from donors’ blood. After three weeks, the valves were ready to go into patients. The Cebotari et al paper in Circulation described “the results of 3.5 years of follow-up”.

No follow-up

The Haverich PowerPoint presentation from 2013 shows as evidence for the 8-year follow-up the photographs of two healthy young adults, the then-13 year old girl is now a mother. Their pulmonary valves allegedly grew: their measurements are given with 26 and 24 mm, the size of adults. A medical miracle indeed, dead tissue alive and growing, thanks to German regenerative medicine. An a close reading of the original Cebotari et al paper shows that the authors neglect to mention how big the implanted cadaver valves were to begin with. On the other hand, it is mentioned that the 11-year old recipient patient had a previous “transannular enlargement of the RVOT [right ventricular outflow tract, -LS]”, which might have made it easier to accommodate an adult valve transplant. For the other patient, the 13 year old girl, Cebotari et al write: “In patient A there was a mismatch between native and TE valve annulus size (10 mm versus 21 mm)”. This basically means, the child patient received a pulmonary valve of an adult cadaver. Yet, I was also informed by cardiovascular specialists that adult-size valves are indeed often implanted into patients with smaller hearts after a kind of folding or artery dilation.  It is therefore very likely that Haverich and his team implanted adult pulmonary heart valves into these two children, while later on insinuating that these grew with them. In any case, no further follow-up studies on these two Moldavian patients were ever published since that Cebotari et al 2006 paper. All we have are two their photographs and Haverich’s word.

On the same day in 2006, Haverich published another paper in Circulation, authored by Lichtenberg et al and titled “Preclinical Testing of Tissue-Engineered Heart Valves Re-Endothelialized Under Simulated Physiological Conditions”. The described sheep experiments were unlikely to pre-date the 2002 Moldavian human experiments, since they were first “presented at the American Heart Association Scientific Sessions, Dallas, Tex, November 13–16, 2005”. Macchiarini first transplanted humans, then supplemented animal experiments afterwards, and it seems very much that Haverich also tested the regenerated valves on sheep after they were first tested on humans. In any case, the sheep paper by Lichtenberg et all provides little evidence in regard to bioreactor-facilitated growth or actual revival of decellurised valves. Both donor and recipient merino lambs were between 10 to 12 weeks old, the follow-up study took only 3 months. The authors concluded that the “in vitro re-endothelialization” was beneficial as compared to direct implantation of decellurised valves, since it helped to avoid “thrombotic formations and neointimal hyperplasia” in lambs.

Never change a running system?

Readers might be forgiven to feel confused now. Initially I spoke about dead heart valves mysteriously regenerated inside the patients’ hearts, by their own blood. Then I went on narrating  how Haverich and his team regenerated heart valves in vitro, using blood-derived endothelial stem cells and a bioreactor (prior to implanting them first into two Moldavian children and afterwards into seven German sheep). But please be lenient with me: it was actually Haverich himself, who without any proper explanation switched the technology. He also decided against previously intended “long-term studies” in animals to “validate the benefits of the presented decellularization method as compared with other alternative approaches”. Before the sheep paper was even published, the Hannover regeneration team instead moved straight to recruiting human test patients en masse.

In their next paper in Circulation (Cebotari et al, 2011), optimistically titled “Use of Fresh Decellularized Allografts for Pulmonary Valve Replacement May Reduce the Reoperation Rate in Children and Young Adults”, Haverich and Cebotari described a proper clinical trial on 38 juvenile patients from Germany and Moldova. The young participants (18 recruited at the Cardiac Surgery Center in Chisinau and 20 at the MHH in Hannover) were operated between 2005 and 2010. They were on average 17 years old, but the given range values suggest that some were as young as five, and some as old as 30. In this regard, I learned from a heart valve specialist: “in many cases, 18 year-old patients will accept an adult size heart valve”.

As I mentioned before, the method applied on this large human cohort was suddenly very different: the decellurised valves were implanted without any seeding by any kind of cells or culture in growth medium, in bioreactor or elsewhere. The authors never lost a word about the whys of this peculiar change of protocol.

 

Allegedly, it is Haverich’s decellurisation technique which produces miracles. Other, “pedestrian” heart valves (used in thousands of patients all over the world) remain dead tissue inside a patient and need to be replaced once they wear out or when the heart grows in size. Haverich’s valves however somehow draw stem cells from the recipients’ blood and make previously dead tissue come alive and grow. Except that once again, Haverich and his co-authors failed to provide any real evidence for this. In fact, as the authors write about the source of the heart valves: “the majority of valve donors were adults and, because of this, most of the DPH valves were implanted oversized”.

An extension of this study was published by the Haverich team in 2015, now with Cebotari as last author and for a change in a different journal: Sarikouch et al, Eur J Cardiothorac Surg, 2016. After having implanted the purportedly regenerating heart valves for 10 years into 93 patients, the authors still are not able to confirm their growth. Most patients were adult or near-adult teenagers; almost all implanted valves had adult origin. The implanted adult valves actually shrunk instead of growing, and the increase of pulmonary valve opening (hailed as indication of valve growth) occurred likely because the patients’ originally juvenile hearts grew. All this is evidenced by the authors’ own data.  A cardiovascular specialist I asked for opinion commented:

“The authors claim that fresh valves is less prone to degradation compared to cryopreserved. I think 10 year follow-up is too short to prove that. Interestingly they abandoned stem cell seeding and claim spontaneous seeding.

The claim that the valves will grow seems to have almost no hard supporting data. In fact they use adult “oversized” valves in children and do not define growth. Their reports were about pulmonary valves but I think the real challenge is the aortic high-pressure valve”.

ARISE and shine?

In fact, Haverich and his team already attempted to the implant decellurised aortic heart valves, in the follow-up EU-funded clinical trial ARISE, “on the basis of compassionate use in 34 patients”.  The ARISE project ran from 2008 to 2013 and costed €16 Million, of which EU shouldered €12 Million. As the Consortium website declared:

“The ARISE project will bridge this therapeutic gap [of aortic valve replacements for juvenile patients, -LS] in a Phase II clinical study to determine the feasibility, safety and efficacy of regenerative heart valves for aortic valve replacement”.

Whatever came out of it, Haverich’s aortic replacement valves didn’t seem to have made it into clinical phase 2 trial. The follow-up ESPOIRE focused exclusively on pulmonary valve replacement. Either way, there is obviously there is no evidence for growth of any of Haverich’s heart valves inside his patients. The concept that circulating blood brings those valves back to life, which was never ever observed in any heart valve implanted in the history of this technology, is scientifically shaky at best. The two key arguments why EU and other public funders chose to support Haverich and his business with nearly €7 Million were basically void long before the money started flowing in 2012. On the other hand, Haverich’s valves are much more expensive than any of their established counterparts (see conference discussion with Sarikouch at the end of Sarikouch et al, 2016). Oner wonders what exactly EU’s funding rationale was here.

Moldovan mysteries

Haverich’s initial experiments with Moldavian children were not free from danger to his patients. His use of live donor cells for bioreactor-mediated regeneration was a potential source of severe immune rejection response inside the patients. As a heart valve biomaterials expert commented:

“Pediatric and adolescent patients are certainly more difficult to treat than adults, largely because of their stronger foreign body response and their tendency to scar and fibrose medical device implants”.

In this regard, an article in the German life science magazine Laborjournal from November 2008 offers some potentially important insights (disclaimer: I used to write for several years for Laborjournal and their English-language edition Lab Times; our separation was not friendly). According to the article, Laborjournal was contacted by an anonymous whistleblower with a file about Haverich’s experiments in Moldova and in Germany. Quote:

“The Anonymous claimed, Haverich implanted at least three Moldovan children with cell-seeded heart valves, and refers to a publication by Haverich in Circulation (2006, 114:1132-7) and an article by Bettina Schöne-Seifert in FAZ.Net. The Circulation paper mentions only two patients, in agreement with the statement by the MHH press office”.

So, was there really a third Moldovan child operated by Haverich and his team? If yes, what happened to it? Another quote from the Laborjournal article:

“The Anonymous claimed that the non-seeded heart valve, which Haverich implanted into a German child on September 9^th 2008, was manufactured by Haverich’s company Corlife. Corlife was however only certified for the development of medical products on September 30^th 2008”.

The muddy activities of Haverich’s Corlife

Interestingly, Corlife manager and ESPOIRE consortium member Michael Harder told me something yet again different (his German-language letter, also about Haverich’s commercial conflict of interest with ESPOIRE, available here). Harder wrote to me:

“a. Corlife has since 2013 permission for processing, preservation, storage and marketing distribution of tissue or tissue preparations according to §§ 20b, 2oc AMG [German Medicinal Products Act, permission valid from 30.09.2013, -LS] (File number 41401 / H 137), which has been repeatedly updated and expanded.

b. Corlife has licenses according to § 21a AMG to market decellularized human aortic and pulmonary valves. The document numbers are PEI.G.11634.o1.1 (2013) and PEI.G.11766.o1.1 (2015)”.

As indicated by Laborjournal, Corlife actually received already on September 30th 2008 a certificate for “development of medical devices for cardiovascular surgery” (see certificates here). Yet Harder declared that the first application for the approval was submitted in 2010 and insisted that Corlife did not manufacture any heart valves until 2013. According to him that work was done previously at MHH, and after the permission was obtained, Corlife began to supply heart valves to ESPOIRE. Was the original Corlife valve manufacturing permit from 2008 never used then? Why?

Corlife advertised already in 2010 to be developing and manufacturing decellurised heart valves, with reference to Cebotari et al 2006 (see recovered evidence here and here). In fact, the newspaper Wirtschaftswoche wrote in 2011 that Corlife manufactured heart valves since spring of that year. An article from 2012 in Deutschlandradio reported how cadaver heart valves were decellurised by Corlife and then delivered in sterile wrapping to MHH, where they were implanted by Sarikouch into patients. How long has Corlife been really producing heart valves for in-patient use, is it really only since October 2013 as Harder insists?

Finally, Haverich neglected to declare any conflicts until ESPOIRE began, including in his Cebotari et al, 2011 publication. This despite the fact that his Corlife proudly presents this and his other aforementioned papers in their list of publications. Haverich’s business partner Harder explained to me (his letter here) that when Cebotari et al 2011 was submitted

“it was not predictable if the relevant approval application will be successful. We received a clarifying communication from PEI [Paul-Ehrlich Institute, -LS] only in 09/2011, almost simultaneously with the publication of this paper. Therefore, there was no potential conflict of interest with this publication”

Also, Harder semi-admitted to me that Corlife valves were used in patients for the Sarikouch et al, 2016 paper, though its authors somehow forgot to mention the industrial origin of these valves anywhere. Instead, they hid this important fact by declaring (highlight mine):

“A. Haverich holds shares in corlife oHG, a company for the future processing of decellularized allografts, equivalent to those used in this study”.

EU reviewed and approved to finance the ESPOIRE clinical trial before Haverich’s Corlife received a regulatory agency approval (or the notification of it by PEI) to manufacture the heart valves. In other words, EU gave a private company €1.1 Million, with all competition excluded, while not being sure that Corlife will actually get the permission to manufacture those heart valves. Unless of course, it works circularly: approved EU funding in turn speeds up the regulatory approval machinery.

Part 1 of “Regenerating in Hannover” is available here

---