Bone Marrow Concentrate Treatment Protocol for Knee OA
Bone Marrow Concentrate Treatment Protocol for Knee OA
Consistent with the previously published results from our group and others, significant improvement of pain and function was observed in osteoarthritis knee joints following injection with BMC at both higher and lower doses of nucleated cells. In both groups, most knee joints were diagnosed with early-stage knee osteoarthritis. Although a higher cell count dose was associated with a greater reduction in pain, there were no significant differences in functional improvement including the LEFS and IKDC, and the subjective improvement rating scale following injection with either a higher or lower cell count.
Although absolute difference in pain outcomes between higher and lower doses was less than two points, the minimum important difference defined in the literature; the NPS score in the higher dose group was 50 % lower than that in the lower dose group, a difference which is considered clinically significant especially given the fact that the comparison is made between two low means; therefore we do not expect to see a high absolute difference in scores despite the highly significant difference as a percentage.
Prior studies of cell dose and therapeutic efficacy of the BMC injectate have been inconclusive. In animal studies, intra-articular injection of either BMC (lower MSC content) or culture-expanded MSCs (higher MSC content) facilitated regeneration of damaged cartilage; however, neither approach showed any superiority in therapeutic efficacy. Clinical studies in humans have demonstrated mixed results, with one study indicating the superiority of higher concentrations of MSCs at repair of cartilage and meniscus injury, and another study indicating greater efficacy at lower cell concentrations of culture-expanded allogeneic cells. The findings from the latter study may be explained by a greater host immune response resulting from the higher-doses of allogeneic cells.
Previous work investigating MSC concentration in the reduction of discogenic low back pain, revealed similar findings with patients receiving higher cell concentration intra-discal injections reporting a statistically significant greater reduction in pain compared to lower cell concentrations. The findings of the present study are unique; prior studies have described the use of BMC for treatment of symptomatic osteoarthritis, but there are none that have examined a cell dose response, or identified an optimal threshold count of nucleated cells for maximizing clinical outcome.
Conceptually, it makes sense that a higher BMNC count equates to better pain relief than a lower cell count. Nucleated cell count is a proxy for the total number of MSCs, and MSCs are the cells that contribute to regeneration of intra-articular cartilage. Unlike studies that have focused on culture expanded MSCs, the number of these cells in BMC is much smaller. Hence, dose may be more critical in some patients with poor MSC counts in bone marrow, which is common in older patients. Additionally, BMC constituents including hematopoietic stem cells, T-lymphocytes, B-lymphocytes, monocytes, macrophages, epithelial progenitor cells and platelets, are all capable of producing growth factors and cytokines that together may support a microenvironment that promotes proliferation and functional differentiation of MSCs as well as cartilage repair. For example, co-culture of MSCs with Monocytes has been shown to increase chondrogenic differentiation capacity. Finally, MSCs have been shown to reduce pain in animal models by the release of TGF-beta.
It is important to note that the observed treatment's effect may be attributed to the platelet component of injections; PRP and PL have been shown in numerous studies to improve the symptoms associated with mild-moderate knee osteoarthritis. However, the clinical efficacy of PRP therapy is transient, and relief from pain and function improvement declines to baseline between 6 and 24 months after treatment. Further, the efficacy of PRP therapy is limited in moderate and severe osteoarthritis, versus mild osteoarthritis. Finally, the significant difference in pain score associated with the higher nucleated cell group in this study reveals that varying MSC dose, within a standardized protocol of platelet injections, has an impact on treatment outcome.
Limitations of the present study are typical of registry studies; there was no placebo control group, and there was no randomization of the patients into the lower and higher cell dose groups. As a result it is possible that uncontrolled for confounding factors or even a placebo effect may account for the results. It is unlikely that the positive results observed for both cell count groups were due to response bias; the analysis of the non-responders indicated no significant differences with regard to age, gender, weight, or baseline severity in comparison with the responders. Also the lack of significance of the differences in the functional outcome scales may be due to limited statistical power. Another limitation of this study is that the ROC analysis and ANCOVA models were performed on data subsets obtained from the same population. ROC analysis was performed on subjects who had available baseline and follow-up scores for both LEFS and NPS scales. ANCOVA models included subjects who had available data for all covariates, in addition to the baseline and follow-up scores of the respective outcome scale. Therefore, future research on different populations is needed to confirm our study's findings. Sufficiently powered randomized placebo-controlled trials are needed to validate and expand on these preliminary results.
Discussion
Consistent with the previously published results from our group and others, significant improvement of pain and function was observed in osteoarthritis knee joints following injection with BMC at both higher and lower doses of nucleated cells. In both groups, most knee joints were diagnosed with early-stage knee osteoarthritis. Although a higher cell count dose was associated with a greater reduction in pain, there were no significant differences in functional improvement including the LEFS and IKDC, and the subjective improvement rating scale following injection with either a higher or lower cell count.
Although absolute difference in pain outcomes between higher and lower doses was less than two points, the minimum important difference defined in the literature; the NPS score in the higher dose group was 50 % lower than that in the lower dose group, a difference which is considered clinically significant especially given the fact that the comparison is made between two low means; therefore we do not expect to see a high absolute difference in scores despite the highly significant difference as a percentage.
Prior studies of cell dose and therapeutic efficacy of the BMC injectate have been inconclusive. In animal studies, intra-articular injection of either BMC (lower MSC content) or culture-expanded MSCs (higher MSC content) facilitated regeneration of damaged cartilage; however, neither approach showed any superiority in therapeutic efficacy. Clinical studies in humans have demonstrated mixed results, with one study indicating the superiority of higher concentrations of MSCs at repair of cartilage and meniscus injury, and another study indicating greater efficacy at lower cell concentrations of culture-expanded allogeneic cells. The findings from the latter study may be explained by a greater host immune response resulting from the higher-doses of allogeneic cells.
Previous work investigating MSC concentration in the reduction of discogenic low back pain, revealed similar findings with patients receiving higher cell concentration intra-discal injections reporting a statistically significant greater reduction in pain compared to lower cell concentrations. The findings of the present study are unique; prior studies have described the use of BMC for treatment of symptomatic osteoarthritis, but there are none that have examined a cell dose response, or identified an optimal threshold count of nucleated cells for maximizing clinical outcome.
Conceptually, it makes sense that a higher BMNC count equates to better pain relief than a lower cell count. Nucleated cell count is a proxy for the total number of MSCs, and MSCs are the cells that contribute to regeneration of intra-articular cartilage. Unlike studies that have focused on culture expanded MSCs, the number of these cells in BMC is much smaller. Hence, dose may be more critical in some patients with poor MSC counts in bone marrow, which is common in older patients. Additionally, BMC constituents including hematopoietic stem cells, T-lymphocytes, B-lymphocytes, monocytes, macrophages, epithelial progenitor cells and platelets, are all capable of producing growth factors and cytokines that together may support a microenvironment that promotes proliferation and functional differentiation of MSCs as well as cartilage repair. For example, co-culture of MSCs with Monocytes has been shown to increase chondrogenic differentiation capacity. Finally, MSCs have been shown to reduce pain in animal models by the release of TGF-beta.
It is important to note that the observed treatment's effect may be attributed to the platelet component of injections; PRP and PL have been shown in numerous studies to improve the symptoms associated with mild-moderate knee osteoarthritis. However, the clinical efficacy of PRP therapy is transient, and relief from pain and function improvement declines to baseline between 6 and 24 months after treatment. Further, the efficacy of PRP therapy is limited in moderate and severe osteoarthritis, versus mild osteoarthritis. Finally, the significant difference in pain score associated with the higher nucleated cell group in this study reveals that varying MSC dose, within a standardized protocol of platelet injections, has an impact on treatment outcome.
Limitations of the present study are typical of registry studies; there was no placebo control group, and there was no randomization of the patients into the lower and higher cell dose groups. As a result it is possible that uncontrolled for confounding factors or even a placebo effect may account for the results. It is unlikely that the positive results observed for both cell count groups were due to response bias; the analysis of the non-responders indicated no significant differences with regard to age, gender, weight, or baseline severity in comparison with the responders. Also the lack of significance of the differences in the functional outcome scales may be due to limited statistical power. Another limitation of this study is that the ROC analysis and ANCOVA models were performed on data subsets obtained from the same population. ROC analysis was performed on subjects who had available baseline and follow-up scores for both LEFS and NPS scales. ANCOVA models included subjects who had available data for all covariates, in addition to the baseline and follow-up scores of the respective outcome scale. Therefore, future research on different populations is needed to confirm our study's findings. Sufficiently powered randomized placebo-controlled trials are needed to validate and expand on these preliminary results.
