[slowlexrese]

Age Management Medicine | August 2010
Christopher J. Centeno, M.D.


Several papers have recently been published claiming adult stem cell related mortality and morbidity without the presentation of credible supporting scientific data to show that the stem cells were the causative agent in the injury. We believe this is occurring because many physicians and researchers are not familiar with the basic epidemiologic tenants of determining causation. The purpose of this short review is to demonstrate how causation is being misused and to provide a framework for it's appropriate use as stem cell case reports and studies continue to be published.

Causation is a science unto itself. Seminal works by Hill and Miller have defined key components that are necessary to determine if an exposure caused an illness.[1-2] In this case the exposure is stem cell therapy. These are:

? Temporal Association ? did the condition begin after the exposure and develop at a rate that is consistent with what is known about the condition?
? Lack of Likely Alternative Explanations ? have all other likely causes been excluded?
? Dechallenge ? when the exposure is removed does the condition improve?
? Rechallenge ? does the condition worsen when the exposure is reintroduced?

To better understand the application of these concepts to determining stem cell related risks, two recently published stem cell papers where harm were reported by study authors are useful. The first by Thirabanjasak et al. entitled, ?Angiomyeloproliferative Lesions Following Autologous Stem Cell Therapy? in which the authors described the clinical course and ultimately the demise of a patient with End Stage Renal Disease subsequent to End Stage Renal Disease. The patient received an unconventional therapy consisting of multiple blind injections of autologous G-CSF mobilized CD34+ hematopoietic progenitor cells directly into the abdominal cavity, presumably targeting the kidneys. The injections were unhelpful and the patient began dialysis within a few months. After six months the patient presented with abdominal pain at the site of HSC injection, and subsequent imaging revealed a suspected urothelial cell carcinoma. At 11 months after the stem cell treatment the left kidney was resected. Histopathology revealed multiple benign angiomyeloproliferative lesions at or near the injection sites, which the authors concluded arose directly and/or indirectly from the stem cell injections. The patient ultimately died of sepsis associated with a post-nephrectomy AV shunt insertion. In a simultaneously published opinion piece by Nagy and Quaggin, these editorial authors took the opportunity to speculate that the stem cell treatment precipitated the patient?s demise, despite the fact that Thirabanjasak et al. did not reach the same conclusion. The authors further claimed that clinics in many countries are performing stem cell transplantation procedures on patients with no oversight and little scientific, pre-clinical or medical rationale, but provided no basis or quantification for their claims. Nagy and Quaggin concluded with the assertion that because of this obvious stem cell therapy related death, all stem cell therapies require oversight from the scientific stem cell community and specifically from their organization, the International Society for Stem Cell Research (ISSCR).

Since the editorial authors concluded that the death was likely caused by the stem cells and have proposed that oversight is needed because of this event, actually determining the cause of death using a scientific analysis of causation would seem important. Applying the above categories to the Thirabanjasak et al. paper to determine cause of death:

? Temporal Association ? The symptoms began 6 months after the stem cell re-implant. Since the patient ultimately succumbed to a common complication of dialysis, an AV shunt infection, one could argue that it was the dialysis that ultimately led to her death. The question then becomes, was there documentation of declining kidney function that would draw a temporal association between the stem cell therapy and the need for dialysis? Regrettably, Thirabanjasak et al. fail to document the even the most basic blood work (in this case serum creatinine) to tie the exposure (stem cells) to the need for dialysis. Had they documented a significant and unusual jump in serum creatinine within a reasonable time frame after the stem cell therapy, the case could be made for a strong casual association. However, as it's documented in the Thirabanjasak et al. report, temporal association is weak.
? Lack of Likely Alternative Explanations? In this case the key question is whether the stem cell therapy caused the need for dialysis. Based on the data provided, the most likely cause for the patient beginning dialysis (the event that lead to her demise) was the removal of a barely functioning kidney in a patient with End Stage Renal Disease. Therefore, the reason the kidney was removed becomes the main thrust of a causation determination. The reasons for kidney removal could have been: stem cells caused lesions in the kidney which reduced kidney function to the point that the kidney needed to be resected, the stem cells caused cancer so the kidney needed to be resected to avoid the spread of the neoplasm, or the kidney was removed in error. Reviewing the paper reveals that the first reason to remove the kidney (reduced kidney function) was never documented, so it must be excluded (see discussion above). The second reason (cancer) was thought possible at the time the decision was made to remove the kidney, but later excluded on histology. Since common minimally invasive methods exist to allow for biopsy and histology of a suspected cancerous lesion without removal of the organ (i.e. ultrasound guided biopsy), the suspicion of cancer based on MRI findings would not, by itself, be a valid reason to remove the kidney. Based on the analysis above, the most likely conclusion is that the kidney was removed in error which ultimately lead to the need for dialysis. Hence there is a credible, non-stem cell, alternative explanation for kidney removal and subsequent dialysis.
? Dechallenge and Rechallenge ? Since this was a single treatment, these items don't apply.

In summary, the standards used to determine causation weren't used by Thirabanjasak et al. or the editorial authors to link the stem cell treatment to the demise of the patient. Rather than a scientific discussion of causation, we have a blind leap by Nagy et al. that the stem cells caused the patient's demise. In other words, in this case, the stem cells got a "bad rap".

While the first case involved a mortality, the second case involved morbidity. The second case is a study published by Kishk et al entitled, Case Control Series of Intrathecal Autologous Bone Marrow Mesenchymal Stem Cell (MSC) Therapy for Chronic Spinal Cord Injury. The investigation involved very low doses of non-culture expanded MSC's used to treat chronic spinal cord injury (SCI). The authors ultimately decided that there were no positive effects of the MSC's and that the MSC's caused multiple complications, the most prevalent of which was chronic neuropathic pain in approximately half of the patients injected. If we use the above criteria to review causation for the most common complication, the exposure is intrathecal injection of MSC's and the end result is neuopathic pain. Applying the criteria for scientific analysis of causation would yield:

? Temporal Association ? Kishk et al do describe in their paper on page 5 that "None of the participants complained of neuropathic pain prior to the study. During the course of treatment, 24 of the 43 treated patients (55.8%; P < .001) developed neuropathic pain ranging from 5/10 to 7/10 below the level of their lesion, and this occurred by 1 to 2 days after each injection." This would meet temporal association criteria.
? Lack of Likely Alternative Explanations ? Kishk et al chose an interesting method to deliver cells to the patient, namely serial intrathecal injections which occurred once a month for 6 months. While intrathecal injection of MSC's would be a common sense solution to allow the injected cells access to the SCI lesion, the delivery method itself is invasive. While a single intrathecal injection is common in medicine for delivery of medications or radiographic contrast agents, significant morbidity can occur as a side of effect of the delivery route. As an example, side effects for intrathecal injections were as high as 33% with some techniques in one study.[3] One would have to assume then that the uncommon practice of repeated monthly injections into the intrathecal space would carry a significant morbidity profile itself, regardless of what was being injected. Therefore, while the authors blamed the very high prevalence of neuropathic pain on the MSC's, this disease could have as easily caused by the delivery method injuring partially innervated nerve roots or leading to dural leaks and causing lowered CSF pressure. In addition, the authors fail to document how the intrathecal injections were accomplished, blind or with imaging guidance, the blind delivery route obviously associated with more complications and less accurate placement of cells.
? Dechallenge and Rechallenge ? While this phenomenon is described for another solitary complication (encephalomyelitis), it is not described for the complication of neuropathic pain nor the other more prevalent complications. Since there were serial injections, this test would helped the causation determination, but must be left out.

Many of the complications described by the authors (neuropathic pain, spasticity, sweating below the lesion) are compared to the control group which was untreated. The authors reason that since these signs and symptoms were observed in the treatment group, but not in the control group, then the cause must be the MSC's. However, since the delivery method could have caused any of these side effects, comparison to the untreated control group who didn't receive intrathecal injections would not allow that conclusion.

In summary, the determination of causation requires a structured scientific analysis. While this type of systematic review is common in large, well funded trials, it is often absent in smaller case reports and case series. One concern here is that while adult stem cells may well cause serious complications, neither paper discussed here meets the causation standards to arrive at that conclusion. In essence, both studies jump to conclusions that give adult stem cells, an unjustified "bad rap".


1. Hill, A., The environment and disease: association and causation? Proceeedings of the Royal Society of Medicine, 1965. 5(295-300).
2. Miller, F.W., et al., Approaches for identifying and defining environmentally associated rheumatic disorders. Arthritis Rheum, 2000. 43(2): p. 243-9.
3. Grasso, A., et al., [Side effects of spinal nerve root radiography. How to reduce them?]. Radiol Med, 1990. 80(6): p. 808-13.


Christopher J. Centeno, M.D. is a stem cell expert and founding member of the International Cellular Medicine Society. He practices image guided, percutaneous, orthopedic stem cell therapy in Denver, Colorado, where he maintains a state of the art stem cell culture expansion facility as part of his medical practice. He can be reached at centenooffice@centenoclinic.com.