Graft engineering: how long can you wait, how low can you go, and pandemic readiness

To the Editor:

Ex vivo graft engineering by means of αβT cell depletion is a promising platform to perform an allogeneic stem cell transplantation (allo-SCT), with low incidences of acute and chronic graft versus host disease (GVHD), for matched, mismatched [1, 2], and haploidentical donors [3, 4]. Translation towards broader clinical implementation is facing graft-related variables with unknown impact. Firstly, additional laboratory procedures to perform graft engineering prolong the time from donation to infusion of the graft. Little information is available on how long grafts can be handled until infusion, or on how many CD34⁺ cells must be harvested from a donor to attain a sufficient allograft resulting in profound engraftment. Graft engineering usually requires higher cell numbers, as the engineering process results in partial loss of cells. Finally, the option to cryopreserve an engineered graft remains essential during the time of an global pandemic [5], and more information is needed on whether the freezing procedure of an engineered allograft compromises outcome. Within this context, we analyzed the questions “how long you can wait” until infusion, “how low can you go” with CD34⁺ cell numbers, and whether cryopreservation puts patients at risk for primary graft failure.

All adults who received an αβT cell depleted allo-SCT product between 2013 and 2021 (n = 237) and have provided written informed consent in accordance with the JACIE guidelines and approved by the local ethical committee (METC nr 21-322) at the University Medical Center Utrecht (UMCU) were included in this retrospective analysis (Baseline characteristics, Table S1). Clinical data were reported to and extracted from the EBMT registry. All patients received a uniform myeloablative conditioning regimen [1]. αβT cell reduction was performed as previously reported [1]. In 2018, CD19 depletion was added to the graft engineering protocol [6]. Cryopreservation was performed in 24 patients before (n = 3) or during (n = 21) the COVID-19 pandemic, in line with EBMT and donor center guidelines [7]. Neutrophil recovery was defined as the first of 3 consecutive days with an absolute neutrophil count (ANC) greater than 0.5 × 10⁹/L. Platelet recovery was defined as a platelet count greater than 50 × 10⁹/L, without platelet transfusion in the preceding 7 days. Primary graft failure was defined as cytopenia and marrow hypoplasia at day 28. Secondary graft failure was characterized by loss of donor cells after initial engraftment and recurrent neutropenia and marrow hypoplasia. Poor graft function is defined as severe cytopenia of at least two cell lines in the presence of a hypoplastic bone marrow with full donor chimerism.

Graft composition of engineered products prior to infusion was analyzed by flow cytometry as part of their quality control. For fresh grafts, the median cell numbers were 5.8 × 10⁶ CD34⁺cells/kg (range 1.7–11.1), whereas the CD34 + cell number for frozen grafts, after thawing, was somewhat smaller (median 4.6 × 10⁶ CD34⁺ cells/kg (range 2.6–11.3) (p = 0.009). This is explained by a lower recovery of CD34⁺ cells (Supplementary Table S2). Consistent with this finding, the median number of infused CD3⁺ T cells was also higher in fresh grafts as compared to frozen grafts after thawing; 5.9 × 10⁶ T cells (range 0.5–48.9) vs 4.7 × 10⁶ (range 0.6–13.3) (p = 0.057) (Supplementary Table S2). The median time to neutrophil recovery was 14 days in the fresh cohort (range 4–26), and 15 days in the frozen cohort (range 10–29) (p = 0.015). For platelets, the median time to recovery was also 14 days in the fresh cohort (range 9–440) and 15 days in the frozen cohort (range 11–110) (p = 0.2) (Supplementary Table S2; Fig. S1A, B). This is in line with a recent report for T cell replete grafts that freezing can impact platelet recovery [8].

Dutch guidelines advise for a maximum of 48 h from apheresis to the time of infusion. For 60 of the 237 products (25%) out-of-spec (OOS) procedures were initiated, as they exceeded this recommended time frame. 85% (n = 51) of all OOS were observed in the fresh cohort (Supplementary Table S3). To assess the impact of time from apheresis to either infusion (‘fresh cohort’), or time to cryopreservation (‘frozen cohort’) on primary engraftment, we compared outcomes in patients who received an allograft infused/cryopreserved within 48 h after apheresis (‘short’) to outcomes of patients who received the allograft after 48 h (‘long’). Both for ‘fresh’ and ‘frozen’ grafts, we found no impact of time from apheresis to infusion on neutrophil or platelet engraftment (Supplementary Table S3; Fig. S1A, B). In addition, we observed no impact of prolonged time to infusion nor cryopreservation on the overall low incidence (3%) of graft failure or poor graft function (Supplementary Table S3).

Graft engineering, as well as cryopreservation, can result in a partial loss of CD34⁺cells. This anticipated loss is usually compensated by harvesting more stem cells from a donor. We ordered grafts of 10 × 10⁶CD34⁺cells/kg. 12% of the donors in our cohort underwent an additional day of collection (Supplementary Table S2). Grafts of 7.8 × 10⁶CD34⁺cells/kg (range 2.4–18.0) were received after collection. Median recovery of CD34⁺cells after the completion of the depletion + /− cryopreservation procedure was 78% (fresh cohort) and 65% (frozen cohort) (Supplementary Table S2). Within the EBMT, allografts of more than 4 × 10⁶/kg CD34⁺cells are recommended. We therefore compared outcomes of grafts with cell numbers lower than the recommended 4 × 10⁶ CD34⁺ cells/kg to allografts >4 × 10⁶ CD34⁺ cells/kg. Here we show that for fresh, as well as for frozen allografts, infused CD34⁺ cell numbers did not correlate with neutrophil engraftment (Supplementary Table S3, Supplementary Fig. 1C). We did observe a significantly longer time to platelet engraftment in the recipients of smaller allograft in the fresh cohort (16 versus 14 days, p = 0.007) implying that lower CD34⁺ cell number mainly affected thrombocyte recovery (Supplementary Table S3). Seven cases of either graft failure or poor graft function were observed, all in recipients of fresh allografts of >4 × 10⁶/kg CD34⁺ cells/kg (Supplementary Table S3).

Finally, we analyzed long term outcome of patients receiving fresh versus frozen products, with different time to infusion and lower and higher cell numbers. We could not observe difference in overall survival when comparing outcome of patients receiving CD34⁺ cell numbers below 4 × 10⁶/kg (n = 32) versus above (n = 205) (Fig. 1a), neither between cohorts which had different times to infusion (<48 h n = 177, >48 h n = 60) (Fig. 1b) nor between patients receiving a fresh (n = 213) or frozen allograft (n = 24) (Fig. 1c). Furthermore, there was no significant impact of any of these variables on acute or chronic GvHD (Supplementary Table S4).

Fig. 1. Impact of cryopreservation, time to infusion and cell numbers on overall survival.

Comparison of overall survival in (a) cohorts with cell number higher than (gray) or below (black) 4 × 10⁶/kgCD34⁺cells, (b) cohorts with time to infusion below (gray) and above (black) 48 h, and (c) fresh (gray) versus frozen (black) cohort. Logrank test was used to assess differences between the groups. Bw bodyweight.

In summary, we show for the first time in a αβT cell depletion transplantation platform that infusing stem cell products until 68 h after apheresis is feasible and does not impact long-term survival. Our data also challenge the currently recommended lowest number of 4 × 10⁶CD34⁺cells/kg for allografts [9]. In line with guidelines for cryopreserved autologous stem cell grafts, which advice 2 × 10⁶CD34⁺cells/kg [9], we advocate that 2 × 10⁶CD34⁺cells/kg are sufficient for both fresh and frozen allografts, which could prevent the need for an extra day of apheresis. Conflicting data have been reported on long-term outcomes with frozen products [10–12]. Long-term clinical outcomes after allo-SCT with a frozen product as a consequence of the COVID pandemic are only slowly emerging. Within this context, our data add important information to the field by showing in a real-world data cohort with a reasonable follow up that cryopreservation of engineered grafts is a valid option.

Author contributions

KN, KW collected and analyzed data, JS, AS, AvR, LvdW, LGMD, RO, RR critically reviewed all procedures and data. MdW and JK designed the study. KN, KW, AS, MdW and JK wrote the manuscript. All authors approved the final version of the manuscript.

Funding

Funding for this study was provided by Miltenyi Biotech to MdW and JK, and by the KWF UU 2015-7553 to MdW; KWF, UU 2018-11393, UU 2018-11979, UU 2020-12586, UU 2021-13043 to JK and 2021-13493 to MdW and JK.

Data availability

The data that support the findings of this study are available from EBMT the registry, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are, however, available from the authors upon reasonable request and with permission of the corresponding author.

Competing interests

JK is inventor on multiple patents dealing with gdTCRs, ligands, isolation strategies of engineered immune cells. JK is cofounder and shareholder of Gadeta (www.gadeta.nl). JK and MdW received research, advisor and clinical study support from Miltenyi Biotech. JK received further research support from Novartis and Gadeta.

Supplementary information

The online version contains supplementary material available at 10.1038/s41409-023-01976-8.