A boost for poor graft function

In this issue of Blood Advances, Shapiro et al¹ report the results of a retrospective study evaluating the use of a CD34⁺ selected stem cell boost (SCB) after allogeneic hematopoietic cell transplantation (alloHCT), predominantly for poor graft function (PGF). An overall clinical response (complete response [CR] + partial response [PR]) was achieved in 36 of 58 (62%) of patients and in 69% of patients with PGF. Active infection at the time of SCB was the greatest predictor of poor response and survival following SCB.

PGF following alloHCT is a rare complication after alloHCT, with a median incidence of 7% (interquartile range, 5%-11%; 22 cohorts) reported in a recent systematic review and meta-analysis.² Importantly, although the definition of PGF is heterogeneous across studies, it is usually defined as the development of 2 or 3 cytopenias with full donor chimerism and the absence of disease relapse. Of note, some definitions also include dependence on red blood cell (RBC) and/or platelet transfusions and/or growth factor support. Importantly, PGF is clearly distinct from graft rejection (<5% donor chimerism). PGF is a potentially life-threatening complication after alloHCT because it may lead to serious complications including a high incidence of fatal infections and severe hemorrhagic episodes due to prolonged neutropenia and thrombocytopenia. Furthermore, RBC transfusion increases the risk of iron overload after alloHCT.³ PGF is associated with a 2-year overall survival (OS) of 53% (95% confidence interval [CI], 45%-61%; 23 cohorts).² Patients who do not recover marrow function have a particularly poor prognosis with a 2-year OS of only 6%.⁴ 

Therefore, different approaches have been investigated to manage PGF without the need of a second alloHCT. Use of growth factor support, (eg, with a thrombopoietin receptor agonist [TPO-RA], given its broad effect on hematopoietic progenitors and its ability to stimulate multilineage recovery), has been reported with promising results.⁵ Similarly, bone marrow derived mesenchymal stromal cells from a third-party donor have been investigated and may be a valid alternative when available.⁶ Finally, granulocyte colony-stimulating factor (G-CSF)–mobilized, CD34⁺-selected SCB provides hematopoietic stem cells without an alloreactive T-cell component and has been used as therapy in PGF, with reportedly high overall response rates of 36% to 81% .^7,8 Nevertheless, given the logistic challenges of donor-cell availability, and a lack of proven factors that predict response to SCB, it is probably important to identify such factors to implement this therapeutic strategy for patients with PGF to achieve a clinical benefit.

Therefore, Shapiro et al conducted a large single center, retrospective analysis to evaluate potential predictors of response and OS following SCB. More than half of the 58 patients received a SCB for PGF (n = 32, 55%), whereas 45% did not meet the PGF criteria used in this study, primarily because of incomplete donor chimerism (n = 10, 17%), absence of bone marrow hypoplasia (n = 8, 14%), or presence of measurable nonmorphologic bone marrow disease at the time of SCB (n = 10, 17%). The overall response rate was 69% in patients with PGF and 54% in others, translating to a higher, although not statistically significant, 2-year OS of 59% in patients with PGF compared with 29% in others (P = .08). Of note, the cumulative incidence of 6-month grade 3-4 acute and 2-year moderate to severe chronic graft-versus-host disease (GVHD) rates were 3.4% and 12%, respectively, confirming that infusion of G-CSF mobilized CD34⁺-selected SCB without additional immunosuppressive prophylaxis is safe and does not translate to severe GVHD.¹ 

Of note, among the 15 patients with active bacterial or fungal infection at time of SCB, only 3 achieved an objective clinical response (CR, 2; PR, 1; 20%) whereas 33 of 43 (77%) patients without active infections achieved a PR/CR (P = .0001). Importantly, other parameters such as bone marrow fibrosis, presence of measurable residual disease, cytomegalovirus reactivation, or use of myelosuppressive medication at time of SCB were not associated with treatment failure. Hence, in multivariate analysis, an active infection at time of SCB was the stronger predictor of mortality with a hazard ratio of 3.25 (95% CI, 1.28-8.22; P = .013) for OS. The detrimental effect of active infection is in line with previously published data,^9,10 and may be related to the suppression of hematopoiesis in the setting of sepsis and excessive inflammatory cytokines. These findings strongly suggest that while it may seem beneficial to administer an SCB during active infection (ie, to facilitate neutrophil recovery and infection resolution), this study suggests that such an approach does not provide significant clinical benefit and should be delayed until the active infection is resolved. Furthermore, despite the logistical burden associated with SCB procurement and administration, these results suggest that this treatment should be implemented early, before severe infection develops.

Interestingly, the authors also analyzed the immune reconstitution of T cells, B cells and natural killer cells and did not find any association with treatment response. Nevertheless, it would be important to compare these results with patients who did not develop PGF to gain more insight regarding the interplay between immune reconstitution after alloHCT, development of PGF, and the response to SCB.

In summary, despite being retrospective and with a limited sample size, this study confirms that SCB is an effective treatment for patients who develop PGF after alloHCT, in the absence of active bacterial or fungal infection. Further studies should investigate the respective use of growth factors, particularly TPO-RA and SCB, in these patients. The early evaluation of growth-factor effectiveness may be important to facilitate the early use of SCB before severe infections develop.

Conflict-of-interest disclosure: F.M. reports honoraria from Bristol Myers Squibb, Therakos/Mallinckrodt, Sanofi, Jazz Pharmaceuticals, Gilead, Novartis, AstraZeneca, and Merck, all outside the submitted work. M.M. reports grants, lecture honoraria, and research support from Adaptive Biotechnologies, Amgen, Astellas, Bristol Myers Squibb/Celgene, GlaxoSmithKline, Janssen, Jazz Pharmaceuticals, Novartis, Pfizer, Takeda, and Sanofi, all outside the scope of this work.