Superior physical and mental health of healthy volunteers before and five years after mobilized stem cell donation

A broad panel of immunodeficiencies, bone marrow failure syndromes and refractory hematological malignancies are treated with allogeneic HSPC transplantation [1]. For the majority thereof, G-CSF-mobilized peripheral blood HSPCs are the preferred graft source. Besides originator G-CSF, several biosimilars are available for HSPC mobilization [2, 3].

Biosimilars are biologics matching their reference biologic in terms of safety, efficacy and quality [4]. Similarity needs to be demonstrated through biochemical characterization, biologic activity in vitro, and clinical similarity in a sensitive indication [5, 6]. Once similarity is shown, the reference’s positive risk–benefit profile is assumed for the biosimilar [2, 7]. G-CSF biosimilars are thus approved for most indications for which the reference is licensed, of relevance to this work including HSPC mobilization in patients and donors [2, 8].

Efficacy and safety of G-CSF in healthy donors have been documented [9‐11], including for biosimilar G-CSF [2]. Effects of HSPC donation on long-term physical and mental well-being, however, have not been explored. We therefore specifically analyzed long-term effects of G-CSF on mental and physical health of HSPC donors.

During the study period, 242 healthy donors were included in the study. To mobilize HSPCs, all participants received filgrastim and no more than two aphereses were performed in each donor. Demographic characteristics are presented in Table 1.

Since there is still an ongoing debate if G-CSF may influence on the occurrence of malignancies, we first analyzed the incidence of malignancies in our study population and compared it with the general German population. Although two cases of malignant melanoma and one case each of Hodgkin’s lymphoma and thyroid cancer statistically exceed the expected incidence for a cohort the size of ours (Table 2), post hoc power analysis for the comparison of incidence rates showed that the power is not sufficient to show a significant difference between both groups for this and similar health outcomes (Additional file 1: Table S1).

Next, we focused on the long-term effect of HSPC donation on physical and mental health. Our study is unique among all previously performed donor outcome studies in that physical and mental health is formally quantitatively assessed using validated sociological tools. The SF-12 questionnaire was applied at each visit to measure these factors from the donor's perspective. Similar baseline scores for physical and mental health were observed, followed by a transient dip only in physical well-being after mobilization with G-CSF, assessed just prior to the first apheresis, then a slow but steady decline of both physical and mental health scores over the observation time (Fig. 1A). Higher baseline and 5-year scores were associated with younger age and male sex. Before mobilization and donation, donors, especially younger donors, had significantly higher scores for both mental and physical health compared to two age-adjusted cohorts of healthy Germans (Fig. 1B, D). Higher scores are maintained throughout the observation period irrespective of the age at donation, for a minimum of 5 years (Fig. 1C, E).

Since healthy donors show a transient dip in physical health, we analyzed if the most common AEs at apheresis might be associated with lower physical health scores. Indeed, significantly more strongly reduced physical health scores are observed in donors experiencing bone pain (physical health score in patients with bone pain: 48.2 ± 8.5) when compared to those without bone pain (physical health score in patients without bone pain: 53.9 ± 4.8, p < 0.05). None of the other more common AEs had a significant association with the physical health score (Additional file 1: Table S2). To gauge whether adverse events during the post-mobilization period might affect willingness to provide long-term follow-up and hence, skew our analyses, we analyzed type and frequency of adverse events after the mobilization period in participants who complete the five-year period and those who did not. There was a trend in favor of patients with serious adverse events, i.e. the overall favorable outcome is observed in spite of a selection bias for donors with health events (Table 3). Finally, we compared the SF-12 scores one month after apheresis in donors subjected to one versus two aphereses to ask whether providing two consecutive apheresis was associated with measurably different, presumably lower, health scores. We did not find significant differences in physical (one apheresis: 55.2 ± 4.9, two aphereses: 55.7 ± 1.8; p = 0.682) and mental health score (one apheresis: 54.9 ± 6.6, two aphereses: 54.0 ± 5.5; p = 0.628) between both groups (Additional file 1: Table S3).

We had previously reported that mobilization with biosimilar G-CSF was efficient as well as acutely safe with no unexpected or limiting adverse events [5]. Analyses of 5-year outcomes do not change this impression. Attributable (severe) adverse events outside the immediate mobilization/post-mobilization period were not detected. Potential long-term effects of G-CSF administration, specifically a contribution to cancer, has fueled ongoing debates [17‐19]. Therefore, we analyzed the limited number of malignancies and could demonstrate that attributability is not suspected, the excess in incidence likely random. Further malignancies were not observed, specifically not myeloid leukemias/myelodysplasia a causal association of which with G-CSF could possibly be construed. Clearly, due to the small cohort size the power to detect subtle differences is limited. Our findings corroborate much larger earlier studies supporting long-term safety of healthy-donor G-CSF stimulation [9, 20].

The distinguishing, indeed unique feature of our long-term donor follow-up is the formal longitudinal assessment of physical and mental health using validated sociological tools. The sensitivity of the SF-12 questionnaire to catch changes in physical or mental health in longitudinal analyses is well documented [10, 21]. Our observations confirm this notion for the physical health score: The association of lower scores with the mobilization period is apparent. This transient dip in perceived physical health is likely causally related to prevalent, intervention-inherent adverse events, especially pain that the SF-12 explicitly queries and thus sensitively detects. Before mobilization and donation, donors, especially younger donors, had significantly higher scores for both mental and physical health referenced against two independent age-adjusted cohorts of (self-reportedly) healthy German non-donors [22, 23]. As to the reasons for this observation, we can only speculate. We propose that a significant contributor likely is selection bias, both self-selection, whereby individuals with (perceived) superior physical and mental health may be more likely to register as donors, and selection during donor eligibility testing [12]. Moreover, absent financial incentives for registration as stem cell donor and stem cell donation, the only driving force for stem cell donors is altruism. It is reasonable to assume that such a positive attitude pervades the personality, is not restricted to the act of donation and is a sustained positive risk factor especially for superior mental well-being. We further propose that overall, donors may be more body conscious, take better care of their health including leading, on average, relatively healthier life-styles. A substantial body of work has studied motivations for and well-being of blood donors, a markedly less invasive form of donation. In agreement with our hypotheses for stem cell donors, selection bias (only healthy persons being allowed to become donors [24]) and altruism, as well as the belief that blood donation provides health benefits [25] have been mentioned. Importantly, higher scores are maintained throughout the observation period, for a minimum of 5 years, i.e. G-CSF mobilization and HSPC donation are not associated with accelerated deterioration of physical or mental health. For one, the earlier discussed selection biases likely continue to be relevant, as well as the positive personality traits (the “happy donor” effect) inducing an individual to “share the gift of life” with a stranger are likely sustained. Secondly, we cannot exclude the possibility that the act of donation itself imprints or reinforces positive, health-adaptive attitudes. Indeed, evidence has been provided that positive events such as a positive donor experience have long-term beneficial effects [26]. Finally, the donor experience may have reminded the donor of the value of good health and have induced or re-enforced healthy habits or life-styles, especially since the donors are informed during donor evaluation about the possibility of additional donations for the same patient in the future, so that a sense of responsibility towards “their” patient may further incentivize the donor to maintain a healthy life-style. To ascertain whether the positive health outcomes were not a subject of post-donation selection bias, we additionally compared donors completing or not follow-up with respect to well-being, probability to suffer adverse events or undergoing two-day apheresis. Neither affected the probability of providing long-term follow-up nor baseline SF-12 scores, thus the followed-up cohort is representative of the complete study population and possibly, G-CSF-mobilized donors per se.

The choice and limitations of the reference cohorts also merit discussion. Unlike the heavily selected donor cohort, these represent a self-reportedly healthy, but not nearly as stringently characterized population. In this sense, some of the superior well-being of the donor cohort, especially in the physical domain, should probably not come as a surprise. Another potentially meaningful difference between the donor and reference cohorts is the longitudinal nature of the former, cross-sectional nature of the latter. However, the demonstrable re-test reliability of the SF12 implies that this effect should be relatively subtle. Significant cultural effects of SF12 scores have been acknowledged; this ethno-cultural bias is accounted for by the selected reference populations. The ideal comparator for our donor population, an age- and sex-matched cohort of similarly selected individuals not proceeding to stem cell donation but similarly followed over time, was not available.

Our data corroborate long-term safety of (biosimilar) G-CSF for healthy-donor HSPC mobilization. The unique, previously not performed longitudinal analysis of physical and mental health with validated health assessment scores demonstrates strongly superior health of HSPC donors before and at all times after mobilization, formally underscoring absence of long-term adverse health outcomes after HSPC donation.