Fraud isn't the only reason for retraction

Several papers have been retracted lately because they invovled fictitious data, fictitious study subjects, or the willful suppression of data. My previous posts on the importance of "smaller" ethical lapses and my own request for the retraction of an article for failures to comply with editorial policy, and local and national guidelines emphasize that fraud is not the only justification for retraction. Several failures - the type that might be considered "smaller" sensu the Scientific American editorial - led the American Journal of Bioethics to announce a retraction today:

"For example, Dr. Hwang was found to have used a far greater number of oocytes than originally claimed in the Science article. The KNBB committee also found that a number of egg donors suffered side effects from the oocyte donation process, and concluded that the oocyte donors were neither given sufficient information about the risks of donation nor were they asked to sign the consent forms approved by the Institutional Review Board.

It now appears likely that the oocyte and somatic cell donation consent procedures described by Hyun and Jung, which, they report, were developed through what they described as close collaboration with Dr. Hwang and his team, were not used for donor screening and enrollment in the research that gave rise to the now discredited Science paper (Hwang et al. 2005). Neither Dr. Hyun nor Dr. Jung have been accused of any misconduct with respect to the contents of this paper, but the Editors of The American Journal of Bioethics nevertheless believe that a retraction of their manuscript is warranted. Both authors have agreed to a retraction."

Failure to abide by approved protocols, unanticipated adverse effects not reported in the publication, and using quantities of materials in excess of those approved in protocols and reported in a publication are all examples of "smaller" ethical lapses mentioned in the editorial statement.

Lie about something that did not happen, and all hell breaks loose. But if it's just a cover up...

...that's another story.

Consider the outrage and press coverage of fabricated stem cell lines, fabricated study subjects and the breaking news of a
Japanese scientist suspected of fabricating papers. Witness the freakishly adamant outrage at a drug addict's stories, which *shock* include exaggerations and falsehoods.

Actually doing something wrong, say abusing animals or torturing prisoners, and few even bat an eyelash. Even starting a war under false pretenses doesn't get a response "on scale" with the severity of the act. I'm confused and disappointed.

The Scientist : PETA asks journal to retract paper

The Scientist : PETA asks journal to retract paper

Pretty self explanatory.

Don't Forget About the Little Ethical Lapses: BLOG: SciAm Observations

This SciAm piece was very compelling:

Don't Forget About the Little Ethical Lapses: BLOG: SciAm Observations:

"Do the small but daily compromises do as much damage to science, cumulatively, as the high-profile cases do? It seems entirely plausible. Precisely because these lapses are small, they seldom affect the final results, so they scoot right by science's self-corrective mechanisms. But that doesn't mean they don't distort science. "

And in the case of science in academia, ethical lapses of all sizes and forms distort the experience and education of the next generation of scientists. This is especially dangerous when no ethics training is provided. If the culture at an institution allows for some rules to be ignored consistently and without punishment - like those prohibiting animal cruelty - then the larger message is that rules don't count. Is it any surprise that technicians at UNC were willing to cut the heads off of unanesthetized mice, even though they acknowledged that it was illegal?

Furthermore, the sense of right and wrong that many students and young scientists bring with them can be tainted in a scientific environment where nondisclosure and disdain for ethical standards are the norm. To the extent that their standards of integrity make them stand out, seem different, or not fit in, the rule-bending culture that plagues some institutions also discourages some bright minds and stymies careers of promising, ethical scientists.

Can we afford to satisfy our curiousity at any cost, even if it is cruelty? Does academic freedom outweigh our obligations to uphold the highest ethical standards in research? That's what some seem to be saying.

In their reticence to respond, the scientific community - universities, professional associations, journals and publishers - a relatively small group of scientists who ignore ethical standards outright are causing the public and the government to lose confidence in the very process of science. Transparency was one of the keystones for the long-held justification for self-policing that science has enjoyed for so long. Lies, cover-ups, collusions and the "blind eyes" of colleagues have eroded this privilege.

The intermediate and smaller lapses simply go unnoticed - but at what cost?

Amid the fury of controversy in recent misconduct cases, many claim scientists will solve the problem of scientific misconduct. Apparently, editors at Scientific American do not agree. Neither do I. When your grants, publications, or ACUC / IRB protocols depend on the cooperation of colleagues and competitors who will review them, how many are willing to blow the whistle, talk about what is questionable or patently wrong or stand up for what is right?

Peer-review isn't a silver bullet; other changes are desperately needed, too. Enron style prosecutions may be the kick in the pants that science needs to get serious about ethical training and obligations.

The straw that broke the monkey's back

Citation: Togawa D, Kovacic JJ, Bauer TW, Reinhardt MK, Brodke DS, Lieberman IH. Radiographic and histologic findings of vertebral augmentation using polymethylmethacrylate in the primate spine: percutaneous vertebroplasty versus kyphoplasty. Spine. 2006 Jan 1; 31(1):E4-10.

Summary of Experiment: Compression fractures are a common and very painful complication of osteoporosis and some traumatic injuries. The two procedures most commonly used to stabilize this type of fracture in people are vertebroplasty and kyphoplasty.

Six elderly female baboons were used to test the efficacy and rate of complications of these two procedures (two vertebrae each). Necrosis, foreign body reaction and leaking in those vertebrae were compared to holes that were created but not stabilized (two vertebrae) and one unaltered vertebrae. Analgesics and IV fluids were administered after surgery. Animals were checked twice per day for “abnormalities.” Three animals were killed “according to AVMA guidelines” 24 hours post surgery, and the other three at six months.

My Notes:

• The efficacy and rate of leaking complications of both procedures among people receiving these same two procedures were tested in a 2002 study, which was not cited. My search revealed nearly 100 papers that compare some aspect of these two procedures in human subjects.


• They fail to mention that baboons aren’t bipedal. The fact that the animals don’t walk upright and would be less susceptible to either compression fractures or vertical compression/stress during recovery has apparently been overlooked.


• These baboons did not have osteoporosis.


• Despite the fact that the entire spine was removed after the animals were killed, they forgot to remove of the vertebroplasty bones for examination.

Quote from the article:

“Unfortunately… bone mineral density of the baboons was much greater than that of human osteoporotic patients so that we were not able to expand the bone tamps to the same degree as is often achievable in humans. This fact may have artificially increased the risk of cement leak in both procedures. Some of the findings in the current study may not be supported in the patients with osteoporotic vertebral compression fractures because of these limitations. Moreover, the baboon vertebrae are morphologically different to human vertebrae (taller height in proportion than human vertebrae), so it is difficult to simulate human osteoporotic vertebral compression fracture. Future studies should attempt to induce more significant osteoporosis in other models using diet, pharmacologic, and surgical (i.e., oophorectomy) means.”

Why is the fact that they didn't have osteoporosis unfortunate?

So the authors admit it’s a crappy model, and everyone knows that there is a sea of human research data, but these people want to do more experiments on monkeys anyway?

Author Affiliations:
Cleveland Clinic Spine Institute, Cleveland, OH
Departments of Orthopaedic Surgery and Anatomic Pathology, Cleveland
Clinic Foundation, Cleveland, OH
Department of Orthopaedic Surgery, University of Utah, Salt Lake City, UT.

Funding acknowledged: Funds from “Kyphon Inc.” This competing interest was declared.

Sick and tired of being sick and tired

Citation: Vulliemoz NR, Xiao E, Xia-Zhang L, Ferin M, Wardlaw SL. Melanocortin modulation of inflammatory cytokine and neuroendocrine responses to endotoxin in the monkey. Endocrinology. 2006 Jan 12; [Epub ahead of print]

Summary of Experiment: Primates were sedated and placed in primate chair overnight. Blood was drawn at 8 am and then saline or one of three experimental agents was injected. At 10 am, animals were injected with enough E. coli to induce a significant immune response and “sickness behaviors.” Blood was taken every 30-60 minutes for the next 7 hours. This experiment was performed 3 times with 3 weeks recovery between sessions. Immune and stress responses were measured. Baselines were taken by performing the procedure without the E. coli injection (x3). Immune and stress responses were measured. Short term reaction was studied by putting primates in chairs the morning of the experiment (0800), given saline or experimental agent for a longer period (until 1300), then given E. coli and monitored until 1800. This experiment was performed 3 times with 3 weeks recovery between sessions.

Some of the agents affected aspects of the immune response that increase and decrease inflammation. The way the stress system (The “HPA axis”) responds to these inflammation messengers was also affected.

My Notes:

• There is no mention of managing the “sickness behaviors” in these monkeys. E. coli can be horrid. Symptoms include intense abdominal pain, bloody diarrhea, fever, loss of appetite, vomiting and dehydration.
• In the discussion, they address the fact that animals used in both parts of the experiment responded differently to phase two than “fresh” animals. The drugs tested could have long-term effects on the immune or hormone systems, or interact with one another when one agent is administered after the other.
• Authors mention that their results mirror rodent studies – Wardlaw also does rodent experiments. It's not clear why they are repeating work in primates.

Author Affiliation: Departments of Medicine and Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, NY 10032

Funding Acknowledged: NIH Grant MH55708 (Looks like ~ $1.7Mover five years)

Toxic relationships

Citation: Gestational and lactational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin affects social behaviors between developing rhesus monkeys (Macaca mulatta). T Negishi, H Shimomura, T Koyama, K Kawasaki, Y Ishii, S Kyuwa, M Yasuda, Y Kuroda, and Y Yoshikawa Toxicol Lett, Jan 2006; 160(3): 233-44.

Summary of Experiment: Forty-nine pregnant monkeys were injected with different doses of dioxin every thirty days from the third week of pregnancy until 3 months after delivery. From 1-2 ears of age, the babies performed tests to assess learning ability, social skills and response to novel and stressful events. Social skills were the only variable affected, and differences decreased over time. The dioxin monkeys seemed not to be afraid of things that elicited fear in normal monkeys. This same phenomenon had already been documented in earlier papers (Schantz AND Bowman, 1989; Schantz et al., 1992). They claim that the differences between those results and their own were due to differences in the behavioral tests used.

My Notes:

• Dioxin is the most potent animal carcinogen ever tested, as well as the cause of severe weight loss, liver problems, kidney problems, birth defects, and death. We already know that dioxin causes damage to development, reproduction, and the immune and endocrine systems at infinitesimally low doses (in the low parts per trillion). Toxicological studies have not been able to establish a "threshold" dose below which dioxin does not cause harmful effects.


• The authors indicate that monkeys may be able to overcome the toxic effects of dioxin during development, because some social behaviors changed. However, it’s equally possible that the initial effect was due to novelty, and continued tests with novelty may have revealed persistent effects.


• Experiments were done at an SNBL facility. Female are housed individually most of the time, but with a male 3 days per month if they are not already pregnant.
  Babies were weaned at 13-14 months (much better than 5 months of age in the prior test), but thereafter were housed individually.


• Some of the methods were funky – they didn’t count non-responses to the test environment as a “trial,” which could lead to habituation, etc. In some tests, they are using groups of 3 and 4 monkeys, in others 28…you can’t just pick and choose like that and then argue that the statistics mean anything.


• Their conclusions are wacky:
  

  “However, the results might serve as a warning to humans of potential alterations by perinatal exposure to TCDD in developing social behaviors, and these alterations might not be necessarily considered as apparently negative effects such as learning disability and psychological disorders.”

Author affiliations: University of Tokyo

Funding acknowledged: Core Research for Evolutional Science and Technology (CREST) program of the Japan Science and Technology Agency.

Conservation of lemurs and other endemic wildlife in Madagascar

I think it's wonderful that President Marc Ravalomanana and several motivated NGOs have defined a vision for conservation in Madagascar. With most of the island's forest already gone, and many habitats in need of protection, it is an absolutely urgent situation. However, most recent news announcements out of Madagascar are about mining: gold, platinum, other metals, and even diamonds. Large scale mining, and its purported economic benefits, and biodiversity conservation just don't mix.

Sadly, I think these conservation efforts may be a day late and a dollar (or ariary ) short!

Smoking gun

Citation: TA Slotkin, KE Pinkerton, and FJ Seidler. Perinatal environmental tobacco smoke exposure in rhesus monkeys: critical periods and regional selectivity for effects on brain cell development and lipid peroxidation. Environ Health Perspect, Jan 2006; 114(1): 34-9.

The authors begin their article by citing sources that acknowledge the harm caused by “Environmental Tobacco Smoke,” i.e. second-hand smoke. The sources are 15-20 years old. Since we already know this, it was critical to perform an experiment on pregnant and nursing mother monkeys and their children.

Summary of Experiment: 15 pregnant mothers were obtained (by Duke from CaNPRC). An additional “set” of 16 monkeys was used in a second set of experiments. The 15 dams were assigned to one of three groups: breathing filtered air, breathing ETS pre- and postnatally, and breathing ETS postnatally only. Smoke from the burning end of cigarettes was “aged” and pumped into chambers that housed two monkeys for six hours each day, five days a week beginning 50 days after conception, and for both mother and infant from birth to 13 months of age. Mothers were removed from the enclosure when the babies were “weaned” at 5 months of age. For monkeys breathing only post-natal smoke, the smoke experiment began at 6 mos. At 13 months, monkeys were given ketamine and then overdosed on phenobarb.

There were no differences between groups in weight, growth, general health or “activity” pre- or postnatally. There were significant difference in the number and size of brain cells in some parts of the brain and in how tissues used oxygen.

My Notes :
These results were nothing new – they are highly duplicative of the authors own work:

“Perinatal or postnatal ETS exposure elicited two characteristic patterns of neural cellular effects, both of which resemble earlier findings for effects of prenatal nicotine exposure in rodents (Levin and Slotkin 1998; Roy et al.1998, 2002; Roy and Sabherwal 1994, 1998; Slotkin 1998, 2004; Slotkin et al. 1987b).”

And an ocean of evidence about the harmful effects of smoke in general. Despite this, they want to keep doing ETS experiments on monkeys:

“These neurochemical inferences point to the need for detailed, quantitative morphologic investigations of ETS effects on primate development paralleling those done for nicotine in rodent models (Roy et al. 1998, 2002; Roy and Sabherwal 1994, 1998).”

They favor the use of diagnostic methods available for studying and monitoring the effects of smoke on human children (available for at least ten years):

“…supports the use of nicotine metabolite measurements in fetuses and children as an appropriate predictor of outcome (Eliopoulos et al. 1996; Fried et al. 1995; Jauniaux et al. 1999; Kohler et al. 1999; Ostrea et al. 1994).”

Stop the presses!

“Translated to human ETS exposure, this finding points out the importance of reducing the exposure of young children to tobacco smoke in the home or in child care settings.”

You mean we shouldn't be giving them ciggys OR smoking around them? I thought the issue was smoking around food or office equipment!

Author Affiliations: Duke University Medical Center. Monkeys provided courtesy of CaNPRC

Funding Acknowledged: Research was supported by grants from the Philip Morris External Research Program and the National Institutes of Health (ES011634, ES05707, and RR00169).

The root of all evil

Remember the horrendous bilateral leg fracture experiment from last week? The senior author on that study is featured again this week! I think we should name this guy the Marquee du Sade of primate experimentation. This week he’s published an article about the interactive effects of tooth infection, root canal, bone healing and an experimental bone growth support treatment. They subjected 6 crab-eating macaques to dental hell for approximately one year.

Citation: Bergenholtz G, Wikesjo UM, Sorensen RG, Xiropaidis AV, Wozney JM. Observations on healing following endodontic surgery in nonhuman primates (Macaca fascicularis): effects of rhBMP-2. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2006 Jan; 101(1):116-25.

Summary of experiment: All monkeys were given tooth infections in their upper (4) and lower (4) incisors using a gruesome technique developed 25 years ago:

“The pulp tissue in each tooth included in the study was accessed, lacerated with endodontic files, and exposed to the oral environment for 1 week to induce a pulpal infection.”

A temporary filling was put in each tooth for 4 (upper teeth) or 7 (lower teeth) months. Bone lesions developed in 37/48 teeth. Several months after the temporary fillings were put in, they performed root canals for the infected teeth. Teeth on one side of the mouth were treated with the experimental agent. Dental exams and photos were taken pre- and post-surgery and at 1, 3 and 4.5 months post-surgery. Animals were euthanized for histopath.

There was some degree of healing over time – seen as early as one month after surgery.

“The findings corroborate previous observations on bone healing and cementum regeneration following endodontic surgery including a human biopsy material and experimental lesions in laboratory animals.”

However, there were complications. At 4.5 months post surgery, all animals had completely or partially necrotic pulp in the infected teeth. There were complications in 3 surgeries – which authors attributed to the difficulty of there being two root canals so close together on roots that were narrow and curved (not really very much like human teeth). At 1 month, x-rays revealed that 4 of the canals had been over-filled with root filling material – something that hadn’t been noticed in the post-surgery x-rays.

My Notes:

• Animals were anesthetized with telazol for X-rays and exams
• A prophylactic antibiotic injection was given within one hour of each surgery + 5 day treatment with broad spectrum antibiotics. (Given the necrosis and lesions, not enough to control infection.)
• Long-acting opiate pain killer (buprenorphine) was given for TWO DAYS after each surgery for pain control. AFTER you have holes drilled in 8 teeth with pulp laid bare? Or after 4 simultaneous root canals????? 2 days of pain medicine?
• Euthanasia by Phenobarbital overdose

The experimental agent didn’t have any effect on outcome, but they didn’t say it that way:

“In conclusion, from a clinical perspective this study does not generate support for the use of powerful biologic agents to enhance bone healing following endodontic surgery.”

Um, if there was no difference, the substance didn’t enhance healing and apparently wasn’t powerful.

Author affiliations:
Göteborg University, Göteborg, Sweden
Temple University School of Dentistry, Philadelphia, PaWyeth Research, Cambridge, Mass

No funding acknowledged

3 Rs: reduction and replacement

The FDA announced new guidelines for drug testing today.

"Concerned about a system in which more than nine out of 10 experimental drugs fail when tested in humans, the Food and Drug Administration issued suggestions Thursday on how researchers can more efficiently evaluate the promise of new laboratory discoveries.

The vast majority of drug candidates fail once they're tested on humans, invariably for safety or efficacy reasons difficult to predict based on initial experiments done in test tubes and on animals. That can waste significant amounts of time and money, slowing the process of developing new drugs."

There are provisions for reducing the amount of animal testing done before human testing can begin and for an expanded roll for earlier invetigational trials in small groups of human subjects. One such approach is microdosing, where people are given very small doeses of a drug to study how the human body deals with the chemicals.

OT: I hate mornings

Waking up is horrible, and it takes hours for me to even feel like a human being. And no wonder. A study released earlier this week suggests that we are half-witted when we first wake. The effects on cognition are more severe than either sleep deprivation or being legally drunk.

Morning grogginess more debilitating than drunkeness or sleep deprivation, according to CU-Boulder study

Since another study found that 7% of workers drink on the job and 9% go to work despite nursing serious hangovers, I suppose that the grogginess and hangovers would be worst in the morning, then the on-the-job drinking kicks in at lunch time. Maybe the lesson is to call customer service in the evening if you really need something done.

Monkeys have it, we don't

Just think of all the monkeys that have been infected with SIV and S/HIV in the last 10 years in the effort (wholly unproductive) to find medicines for people with HIV and vaccines against the disease. Turns out that most people lack some genetic machinery (TRIM5 gene) that macaques have for fighting retriviruses like HIV.

A clue from macaques yields evidence for impaired retroviral defense genes in humans

Help! Our empathy is missing!

After reading the article Cells That Read Minds (New York Times), I was struck by the fact that there wasn't a single mention of the human ability to "feel" for the monkeys in these experiments.

They mentioned the electrodes, but not the restraint chair or the crainotomy needed to bolt a steel chamber to her head to hold those electrodes. They didn't mention that she was alone in a cage, rather than with other monkeys with whom she could use her mirror neurons.

So, I wrote a little letter to the NY Times...since they never seem to publish what I send them, I'm posting it here for good measure.
___

To the Editor,

Your story on mirror neurons and the biology of empathy (Cells that read minds, 1/10) overlooked something conspicuous – our society’s lack of empathy for the monkey strapped into a chair with electrodes stuck in her brain. Despite their bright minds and unique personalities, many primates languish in small laboratory cages where they endure pain, fear, boredom and distress. Mirror neurons also provide vicarious terror when they see other monkeys wrestled from their cages, or injected with drugs, or maybe even killed at the end of an experiment. To see an example of what monkeys in laboratories commonly see, your readers can visit www.covancecruelty.com.

Given all we know about primates, it’s clear that how we keep and use them in labs is replete with ethical problems. If we take a long, hard look at the suffering of laboratory primates, our mirror neurons will tell us it’s time put an end to the cruelty.

3 Rs: Reduction

FDA needs to get with the picture: Bayesian statistical methods could bring clinical (and preclinical) trials into the 21st century.

The authors of this study suggest that a different way of looking at trial design would reduce the number of subjects needed for trials and reduce the risk for those who are used as subjects.

Bayesian statistics overview
Bayesian statistics for the statistically inclined

Shock the primatologist

This study wasn’t published in English, but because it is so freaky, I had to include the abstract! I imagine that this was terrifying and I truly hope that no IACUC would approve such a protocol here.

Citation: [Inguinal lymph nodes of monkeys subjected to 8-days-long immersion in water]
Morfologiia, January 1, 2005; 127(3): 45-7.

Abstract: The morphometric study of cytoarchitecture of the structural and functional components of lymph nodes of male monkeys (Macaca mulatta) aged 4-5 years was performed after 8-days-long immersion in water. Activation of lymphocyte mitotic activity in lymphoid nodules (B-dependent zones) was demonstrated in association with an accumulation of small lymphocytes in them and an increase of macrophage reaction. Accumulation of mature forms of plasma cells was detected in all the structures of lymph nodes. It was also found that 8-days-long immersion in water resulted in remodeling of the vessels of lymph node microcirculatory bed, which was accompanied by changes of the endothelium in blood vessel wall and by disturbances of lymph drainage, that lead to accumulation of lymphocytes in lymph node sinuses.

(Russian language publication)

The eyes have it - Part I

Phototherapy is a common, successful treatment for age-related macular degeneration and other retinal disorders among people. Laser treatment of the affected eye can help patients with such conditions, and the power of the laser treatment and lens employed are calibrated to the unique qualities of each patient’s eye – including axial length. Monkeys naturally have axial lengths shorter than those of people. Despite this, and the fact that there are already human data on this treatment, some still choose to experiment on animals.


Citation: Effect of axial length on laser spot size during photodynamic therapy: an experimental study in monkeys. M Kondo, Y Ito, K Miyata, N Kondo, K Ishikawa, and H Terasaki. Am J Ophthalmol, January 1, 2006; 141(1): 214-5.

Summary of experiment: Authors used some pediatric equipment because of the size of the monkeys' eyes. The authors indicate that they have not treated any patient with an axial length in the range of these monkeys.

“Although we have not performed PDT on patients with very short axial lengths, a reduction of the spot diameter should be considered for such cases.” P. 215

Three monkeys with healthy eyes received the laser treatment. After one week there was an unusually powerful response to the laser, so they cut the eyes out to examine them. Once they figured out how their equations for laser calibration had to be adjusted, they performed the experiment on one additional monkey, cut his eyes out and got the results they expected.

My Notes:

• Procedure tested on these monkeys has already been vetted through a variety of clinical trials in humans.
• Despite existing data on humans treated with phototherapy and their natural variation in axial length, AND the fact that experiments with a model eye have shown that there is an important relationship between axial length and how the eye responds to this specific form of phototherapy, investigators decided to conduct an experiment in monkeys.
• The authors did not indicate whether the animals were killed in addition to having their eyes removed.
• There was very little detail about the methods, as this was a short communication.

Author affiliation:
Department of Ophthalmology, Nagoya University School of Medicine, Nagoya, Japan

Funding acknowledged: Supported by a Grant-in Aid 16591747 (M.K.) and 16390497 (H.T.) from the Ministry of Education, Science, Sports and Culture, Japan