In the previous post, I had focused on the implications of the FTC Order issued last October as it pertains to the type of clinical studies required if a clinic or health care provider wanted to make health-related claims about the therapies they provide and to advertise these claims. The FTC obtained Federal court endorsement of a standard in which only results from Level 1, double-blinded, placebo-controlled clinical studies performed by recognized experts in a pathology could be relied on to support making health-related claims about a particular regenerative medical therapy. There are a number of issues with this requirement, which is especially acutely problematic for studies involving BMC. As a starting point, I will illustrate aspects of the BMC-Level 1 problem by continuing with my review of the Shapiro, et al. study (2016) I touched on in the previous post.
The basic design of the FDA-blessed clinical study presented in Shapiro, et al. (2016) is as follows:
- Patients suffering from bilateral knee osteoarthritis were enrolled in the study
- The knees of the patients were treated with either saline or BMC
- The knees were randomly assigned either to receive saline or BMC, but the patient was blinded as to which knee was injected with saline or BMC
- Visual Analog Scale (VAS) and Intermittent and Constant Osteoarthritis Pain (ICOAP) questionnaire were used to track patient-reported outcomes
The authors rated the protocol as a Level 2 study; not quite up to the new standard of Level 1 studies the FTC got the Federal court to endorse last October. On the other hand, the authors also indicated that their study was the first such study to be “FDA-monitored”.
Since each patient received both the saline placebo and their BMC, the authors were able to use each BMC drawn to treat a study subject, so no one was asked to undergo an aspiration, and have it thrown away. In other words, the patients were their own controls, which definitely isn’t a characteristic of a Level 1 study. The results of the six-month milestone comparison showed that there was no statistically significant difference between any of the metrics the patients reported for the knees receiving the placebo versus the BMC treatment. I should point out that patients experienced a statistically significant improvement in pain and quality of life metrics when comparing pre-treatment scores to their results at the 6-month milestone, but there was no statistical difference between the knees receiving saline or BMC at that milestone.
The authors entertained a number of possible hypotheses to explain the lack of a statistical difference between the treatments:
- BMC-treated knees felt so much better that there was a sympathetic biomechanical boost to the saline-treated knee
- The body’s response to the physical act of injecting each knee was sufficient to achieve the observed pain relief
- There was a “strong” placebo effect, since the patients knew that their knees were to receive both saline and BMC
- The BMC treatment in one knee had a systemic impact that resulted in pain mitigation in the saline-treated knee involving cell migration via the peripheral circulation
The authors dismiss the sympathetic biomechanical benefit due to the rapid on-set of pain relief in the saline-treated knee. They indicate that others have reported pain mitigation with a variety of materials being injected, but the durability of those injections didn’t match their 6-month results. The strong placebo argument might have some validity, since they couldn’t rule it out, given that each patient received both treatments as a part of the study protocol. The authors offered up several points in favor of their final hypothesis, including citing evidence that progenitor cells will home to sites of inflammation, and that it is possible that the progenitor cells injected in one knee could migrate via the circulation to the saline-injected knee to help mitigate pain.
In addition to the above listed potential explanations suggested by the authors to explain their curious results, I offer my own set of possible explanations:
- Saline was used as the so-called placebo
- The injection volume was 15 mL for either BMC or saline
- Some of the patients reported a value of 0 (no pain) on some of the metrics for their knee pain prior to treatment
- The original BMC volume was 5 mL, but was diluted to a final injection volume of 15 mL with autologous PPP
With respect to the authors’ speculation on the migratory route taken by progenitor cells from the injected knee to get to the placebo knee, I suspect that it is entirely possible that cells could migrate via the lymphatic system, and to have exosomes play some role in pain mitigation in the placebo-injected knee.
I covered why saline might not be a valid placebo in a couple of posts last year, but suffice it to say that saline injections have been shown to provide analgesic relief for certain kinds of pain, which might include OA-induced pain.
More importantly, the authors injected 15 mL of either BMC or saline into the study subjects’ knees, which I would argue is more of a mini-lavage than an injection. It has been reported that just lavaging OA-afflicted joints can provide pain relief. This explanation might account for the rapid on-set of pain relief in the saline-injected knees as reported by the authors. As indicated above, some of the reported ranges in pre-treatment pain subscores included “0”, meaning that there was no pain reported for that subcategory of pain. I can’t imagine a study subject reporting a score better than 0 following the treatment, so the subjects’ self-reported pain profiles provided an additional layer of confounding influence on subsequent pain assessments.
Finally, the authors diluted the concentrated BMC by 10 mL to get the final mini-lavage volume of 15 mL. Using a substantially diluted BMC preparation is in contrast to evidence presented by Dr. Philippe Hernigou (Chief of Orthopedic Surgery, Henri Mondor Hospital, East University, Créteil, France) in a number of publications in which therapeutic benefit was correlated to the concentration of progenitor cells in the injectate, suggesting that the higher the MSC level the more likely the patient reached a clinically-objective positive outcome (e.g., union from a non-union, intact rotator cuff at 10-year follow-up following rotator cuff surgery with BMC augmentation, etc.). Consequently, it may be that the use of a substantially diluted BMC preparation resulted in a diminished therapeutic benefit.
A follow-up publication by Shapiro, et al. was published last year, so I will review the 1-year outcomes (more of the same, I am afraid), as well as explore in more detail how BMC therapies might be studied in formats other than the FTC/FDA-favored Level 1 clinical study in the next post.