Bone marrow derived CD 34+ cells and leukocytes in 729 children and adults with non-malignant diseases

Background: To our knowledge, no large studies reporting bone marrow (BM) derived cell counts of children and adults with non-malignant diseases have been published so far. Thus, the primary objective was to evaluate BM-derived CD34+ and leukocyte cell counts in 729 female and male patients of different age groups who underwent autologous cell-based therapy for non-malignant diseases. Methods: For this study, a retrospective data analysis of laboratory parameters including BM-aspirated and post-centrifuge concentrated CD34+ cells and leukocytes was performed. Associations and differences of cell counts between age groups, gender, and diagnose related group were evaluated. Results: Included were data of 187 female and 542 male patients aged between 2 and 75 years, who underwent autologous cell-based therapy for various non-malignant diseases. The median percentage of CD34 + cells of leukocytes was 1.10% in BM-aspirate (BMA) and 0.96% in post-centrifuge BMA concentrate (BMAC). Signicant moderate positive correlations were observed between CD34+ cells (count/µl) and leukocytes (count/µl) in BMA and in BMAC. Signicant strong negative correlations were detected between age (years) and CD34+ cells (count/µl/kg) in BMA and in BMAC. No signicant differences regarding CD34+ cells (count/µl) in BMA were detected between adults, while signicant differences regarding cell counts were detected between diagnose related groups, but not between females and males. Conclusions: This study demonstrated BM-derived CD34+ and leukocyte cell counts in BMA and BMAC of 729 patients with various non-malignant diseases. While BM-derived CD34+ cells were signicantly higher in younger patients, similar cell counts were detected within adults.


Introduction
In the past decade, bone marrow (BM)-derived cell-based therapies have been performed in various elds of regenerative medicine. In spite of the fact that BM contains hematopoietic stem cells, the CD34 + cells are an essential part in therapy and are known as indirect indicators for progenitor cells and thus, are considered as the gold standard for stem cell quanti cations. 1 Yet, little is known about standard values of these cells in the BM. CD34 + cell counts are still not routinely reported in therapeutic studies with autologous BM. Thus, crucial knowledge of what we are doing in regenerative medicine is at least in part lacking. The reason for this discrepancy may be that hematooncologist, who have the best knowledge in BM physiology, rarely engage in regenerative medicine, while on the other side, specialists (mainly surgeons) engaged in regenerative medicine have little access to hematology.
Flow cytometry analysis for quantifying single-cells such as the hematopoietic surface marker CD34 has been widely used in clinical practice, especially in peripheral blood of patients with malignant diseases. 2 To our knowledge, comprehensive studies reporting on typical amounts of CD34 + cells in BM are sparse.
Sutherland et al. 3 stated that approximately a range of 1 to 3% of BM cells express CD34 + antigen (without reporting a sample size) and Brooimans et al. 4 detected 1.4% of CD34 + in BM-aspirate (BMA) in a small series of patients.
Stem cell counts are important, since it is assumed that better clinical results can be achieved with a higher donor-site stem cell count. 5 The volume of a graft required for hematopoietic stem cell transplantation is commonly based on either the number of CD34 + cells or the number of total nucleated BM cells. 6 Furthermore, the impact of age is well documented for cell-based procedures, but controversial. On the one hand, grafts from older donors did not adversely affect outcomes of allogeneic hematopoietic cell transplantation as compared to grafts from younger donors in hematooncology, while on the other hand, lower CD34 + cell counts were reported in older donors, 7-9 and older age correlated with inferior results 10 or lower CD34 + cell counts in some speci c diagnoses. 11 Thus, it seems clear that BM cell counts are deteriorating with increased age. However, the majority of studies reported data of patients with malignant diseases or their healthy donors, [6][7][8]10 but little is known about patients with non-malignant pathologies.
With the present study, we are reporting counts of CD34 + cells and leukocytes in BMA and centrifuged BMA-concentrate (BMAC) in a large number of patients who underwent autologous stem cell transplantation for various non-malignant diseases. Cell counts were analyzed regarding in uences of age, gender and diagnose related groups.

Material And Methods
For this retrospective study, all laboratory data of patients who underwent stem cell transplantation for non-malignant diseases up to 2019 were evaluated. The respective diagnoses of patients treated at the institute for regenerative medicine were spinal cord injury, amyotrophic lateral sclerosis, autism, cerebral palsy, neurodegenerative diseases (e.g. multiple sclerosis, Parkinson, ataxia), muscle dystrophy, and neurological diseases (e.g. traumatic brain injury, stroke). Included were female and male patients aged between 2 and 75 years with available BM-derived CD34 + cells (count/µl and count/µl/kg body weight) and leukocytes (count/µl and cells/µl/kg body weight) of the BMA and post-centrifuged BMAC.
Percentages of CD34 + cells of leukocytes and concentrations of BMAC of BMA were calculated. Data from procedures with missing or invalid cell counts were excluded.
The study was approved by the ethics committee of the Medical University of Graz (31-152 ex 18/19, Ethikkommission der Medizinischen Universität Graz, Auenbruggerplatz 2, 1.OG, 8036 Graz, Austria / EU). All methods were carried out in accordance with relevant guidelines and regulations.
All procedures were performed as point-of-care method in a laminar air ow operating room. Aspiration of BM was performed by the same experienced surgeon (G.S.K.) from the posterior and anterior iliac crest. Stem cells were harvested with a Yamshidi Needle (15ga x 2.688in MAX Bone Marrow Aspiration Needle, ARGON Medical devices, Athens, USA) under sedoanalgesia. BMA was retrieved using 10ml syringes and changing direction repeatedly as published by Oliver et al. 12 . Following, BMA was processed in the operating room according to the SmartRedux protocol (Biosafe, Eysins, Switzerland) using a fully automated cell separator system (Sepax® S-100; Cytiva Europe GmbH, Vienna, Austria).
One ml of the total BMA sample was immediately transferred to the same laboratory and analyzed with uorescence activated cell sorter (FACS) using a stem cell kit from Beckman Coulter and the ISHAGE protocol (https://www.bc-cytometry.com/PDF/DataSheet/IM3630.pdf).

Statistical analysis
Demographic details were presented using descriptive statistics. Data distribution was assessed by visual inspection of histograms and the Kolmogorov-Smirnov test. Qualitative data were expressed by numbers and percentages and quantitative data as means with standard deviation or median with range. CD34 + concentrations (in times) after centrifugation were calculated as CD34 + cells (count/µl) in BMAC divided by CD34 + cells (count/µl) in BMAC. For continuous and normal distributed data, independent ttests were applied and Mann-Whitney U or Kruskal-Wallis tests were used for non-parametric data to determine differences between two or more groups, respectively. Bonferroni adjustments served for multiple testing. Spearman's rank correlation coe cients (rho) were used to assess the correlation between parameters. Statistical signi cance level was set at P < .05 (2-sided). Statistical analyses were performed using SPSS Statistics 25 (IBM Corporation, Armonk, NY).

Results
A total of 729 laboratory datasets were evaluated. There were signi cantly more children (n = 445, 61.0%) than adults (n = 284, 39.0%; p < 0.001) and males than females (p < 0.001) treated with autologous stem cell transplantation. Frequencies of CD34 + cells (count/µl) and leukocytes (count/µl) in BMA are presented in Supplement le Figure S1 and S2. The median percentage of CD34 + cells of the leukocytes in BMA as well as in BMAC was 1.1%. All demographic data are presented in Table 1. Very strong and strong negative correlations were detected between age (years) and CD34 + cells (count/ µl/kg) in BMA ( Fig. 2; rho = -0.827, p < 0.001) and in BMAC (rho = -0.712, p < 0.001), respectively. The negative relationship between age (years) and CD34 + cells (count/µl) in BMA (rho= -0.601; p < 0.001) was moderate and weak in BMAC (rho= -0.285; p < 0.001). All correlations between age and cell counts are presented in Supplement le Table S1.
Comparisons between age groups, gender, and diagnose related groups regarding BMA and BMAC showed signi cant differences between children and adults and between diagnose related groups, but not between females and males (Supplement le Table S2).
Percentages of CD34 + cells of the leukocytes in BMA of children and adults according to their nonmalignant diseases are presented in Fig. 3.
Density gradient separation using a centrifuge-based system showed that the lower the CD34 + cells (count/µl) in BMA, the higher the concentration (times) of CD34 + cells (count/µl) in BMAC after centrifugation (Fig. 4)

Discussion
The present study investigated leucocytes and CD34 + cell counts in BMA and BMAC in the largest number of subjects with non-malignant diseases so far and included all ages ranging from 2 to 75 years. One major nding of 729 cell counts was that approximately 1% of leukocytes were CD34 + cells in BMA and in BMAC with slight variations according to age. Second, BM-derived CD34 + cells were signi cantly higher in children as compared to adults, whereas comparable cell counts were detected within the respective age groups of adults. Cell counts were independent of gender.
Results of a previous study reporting only a weak inverse relationship (r=-0.4) between age and the number of CD34 cells seem opposed to our ndings on the rst site. However, they mainly included a small series of elderly patients with chronic heart failure. 11 Indeed, we not only reported a mix of DRGs, we also showed differences regarding various age groups in a much larger sample size. Furthermore, our ndings (higher cell counts in younger patients) support a previous study with a smaller sample size. 6 Autologous BM-derived cell-based therapies in regenerative medicine are on the rise. While it is generally known that BMA contains a mix of nucleated cells and other biologics such as growth factors and exosomes 13 , comprehensive information regarding normal ranges of leucocytes and/or CD 34 + cell counts in BMA and BMAC in a large series of patients of all age groups was missing.
We found a signi cant positive linear correlation (CD34 + = 8.3*10 − 3 leukocytes + 61.2) between CD34 + cells and leukocytes (rho = 0.687; p < 0.001). This nding add further knowledge to the ndings of a study by Terstappen et al. 14  Stem cells are mononuclear cells and therefore, a fraction of the bone marrow derived leukocytes. Surprisingly, to our knowledge no prior study described the positive and strong correlation between both parameters, which might be of use in future clinical practice. Since stem cells can only be identi ed using speci c CD antigen sets (CD34, CD90, CD45, CD107,…), which is costly and laborious, this correlation can be utilized, to predict the amount of stem cells based on the number of bone marrow derived leukocytes alone, which is much easier. As a matter of fact, only a negligible share of publications reports the absolute number of stem cells used per patient. 15 Using the newly described correlation, a much cheaper possibility exists, to assess, if a speci c patient has a high or low stem cell number. It has to be assessed in the future, if patients with a higher leukocyte-and stem cell -count will have better outcomes and might therefore be better suitable for stem cell operations.
According to our data stem cell counts do not deteriorate in adults (18-75 years); this supports the work by Povsic et al. 16 who reported that ageing is not associated with BM-resident progenitor cell depletion (18-85 years).
Yet, we are unaware of any study reporting data of cell counts in children (2-18years). Despite the fact that cell counts of our large group of children had several different diagnoses, leucocytes and CD 34 + cell counts were signi cantly lower in adults. It remains speculative, if the higher cell count in children is necessary for growth and differentiation up to the age of puberty. However, since cell counts remained stable in adults (18-75years), increasing age might not necessarily be a contraindication for autologous stem cell therapy.
Furthermore, lower cell counts may also be due to speci c diseases, as for example reported by Terstappen et al. 14 reporting lower stem cell counts to be indicative for unfavorable outcome.
It is also not clear yet, if heterogeneous clinical results are linked to heterogeneous bone marrow derived stem cell counts. One could assume, that a higher yield of stem cells might result in a superior clinical outcome. Anyway, the possible predictive value of stem cell counts needs further investigations, especially linking donor-site cell counts to clinical outcome and thus, to sort out patients with unfavorable cell counts to avoid unnecessary interventions.
Since the onset of regenerative medicine several centrifuges are available on the market for mononuclear cell concentration. 17,18 The concentration factor in our data showed a negative correlation to the number of leukocytes in BMA. Thus, a higher degree of concentration is found in patients with a lower bone marrow leukocyte count. This in part counteracts the dilution occurring in any aspirate higher than 2ml, which is due to the in ow of peripheral blood into bone marrow during the aspiration process. 19 Despite the large sample size reporting cell counts, there are some limitations: Cell counts were quanti ed at an extern laboratory and therefore, measurement accuracy could be in uenced by transport, however serial tests with the laboratory found no signi cant difference between one to 12 hours of transport in ethylenediaminetetraacetic acid (EDTA) tubes. Furthermore, our data are probably not comparable to individuals with other diagnoses than those of our study group. Severity of illness, not reported in this study, might also contribute to cell count variations and hence, can in uence our data. Due to the retrospective study design, we are not able to provide data on other cells (e.g. red blood cell count) nor are we able to compare BM-derived cell counts with peripheral blood cells. Cell counts might also be in uenced by BMA volume, which we did not measure in this study. Further studies of different age groups are needed (a) to evaluate typical BM-derived cell counts of individuals with other diagnoses than ours, (b) to report cell count related outcome, and (c) to link cell count data to epigenetic variables (e.g. smoking habits, body mass index, etc.) However, this large study provides baseline data for further evaluations regarding autologous BM-derived cell therapies.

Conclusion
This study demonstrated BM-derived CD34 + and leukocyte cell counts in BMA and BMAC of 729 patients with various non-malignant diseases. While BM-derived CD34 + cells were signi cantly higher in younger patients, similar cell counts were detected within adults.

Declarations
The protocol for this retrospective study, including a waiver of the informed consent requirement, was approved by the ethics committee of the Medical University of Graz (31-152 ex 18/19, Ethikkommission der Medizinischen Universität Graz, Auenbruggerplatz 2, 1.OG, 8036 Graz, Austria / EU). All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication
Not applicable.