Peripheral Blood CD34+ Peak Adds a New Layer to Autologous HSCT Assessment
In autologous HSCT, the infused CD34+ cell dose has long served as the standard practical marker of whether a progenitor cell harvest is adequate. This study from a 302-patient cohort argues that the story begins earlier, during mobilisation, where the peak peripheral blood CD34+ count carried independent prognostic value for post-transplant neutrophil recovery.
The finding is clinically plausible. Mobilisation peak reflects how efficiently marrow progenitors are released into the circulation, a process shaped by stem cell reserve, marrow responsiveness, and the biological effect of mobilising therapy. Apheresis yield captures what is collected. Mobilisation peak may capture what the patient was able to mobilise in the first place.
A Cohort Built Around Single-Mobilisation Transplants
Investigators reviewed autologous transplants performed from January 2015 through December 2024 in the Metropolitan HSCT Program of Catania, Italy. Only patients whose graft came from a single mobilisation were included, which is important because mixing products from multiple mobilisations would blur the relationship between the observed CD34+ peak and the final infused dose.
The final cohort included 302 patients, representing 42% of all autologous transplants in that program over the study period. The mean age was 53.2 years, with multiple myeloma accounting for the largest subgroup, followed by non-Hodgkin lymphoma and Hodgkin disease. Mobilisation strategies included cyclophosphamide plus G-CSF, DHAP plus G-CSF, other chemotherapy plus G-CSF, and G-CSF alone. Plerixafor was used in 29 patients, or 9.6% of the cohort.
Neutrophil engraftment occurred in all but four patients, who died of sepsis before engraftment and were censored at last observation. Across the whole cohort, the mean time to neutrophil engraftment was 12.2 days in the abstract; within the results section, the reported cohort mean was 11.5 days among patients given prophylactic post-transplant G-CSF and 13.2 days among those who did not receive it. The difference reached p=0.05.
Post-transplant G-CSF did not erase the mobilization signal. It shortened recovery, but the mobilization biology still mattered.
Infused Dose Mattered, Yet It Did Not Tell the Whole Story
The infused CD34+ dose remained associated with engraftment time. The mean infused dose was 5.1 × 10^6 CD34+ cells/kg, with a median of 4.2 × 10^6/kg and a range from 1.4 to 26.5 × 10^6/kg. When patients were grouped by infused dose, neutrophil engraftment differed significantly across the three strata: 12.8 days in those receiving less than 3 × 10^6/kg, 12.4 days in those receiving 3.1 to 6.0 × 10^6/kg, and 11.4 days in those receiving more than 6.1 × 10^6/kg.
In the Cox model, infused CD34+ dose remained significant when treated as a continuous variable, with a hazard ratio of 1.048. When grouped categorically, both lower-dose strata had slower engraftment than the reference group above 6 × 10^6/kg. In plain clinical terms, more cells still meant faster recovery. But dose alone did not fully explain the timing of marrow rescue.
The Mobilisation Peak Carried Independent Prognostic Weight
The central result came from the peripheral blood CD34+ peak measured during mobilisation, before apheresis. CD34+ monitoring was done daily by flow cytometry beginning 8–9 days after cyclophosphamide-based mobilisation or on day 14 after more intensive regimens such as DHAP. When plerixafor was used, the pre-dose morning value was taken as the peak.
The median mobilisation peak was 120 cells/µL. Patients were then stratified by peak level, and the differences in engraftment were significant across the resulting groups. Mean neutrophil engraftment times ranged from 14.6 days in the lowest peak group to 11.0 days in the highest peak group, with intermediate strata falling between those values. The overall comparison was statistically significant, with p=0.001.
In multivariable analysis, the peak peripheral blood CD34+ count remained significant even after accounting for other factors. The infused CD34+ dose lost that independent status in the final model, while G-CSF use also remained significant. ROC analysis yielded an AUC of 0.695, with a 95% confidence interval of 0.599 to 0.791, for predicting delayed engraftment. For a biologic marker in a real-world transplant cohort, that level of discrimination is modest rather than dramatic, but it is clinically meaningful because it adds information not captured by the infused dose alone.
The model suggests that mobilisation peak is not just another number in the collection log. It is an upstream marker of engraftment biology.
Why Mobilisation Biology May Matter More Than Collection Logistics
Apheresis yield is influenced by venous access, monitoring frequency, procedural technique, and the timing of collection. Mobilisation peak is less vulnerable to those operational variables. It reflects marrow output under a defined mobilising pressure, which is why it may better represent the underlying quality of the graft source.
That distinction has practical value in transplant planning. A patient who reaches a robust peripheral blood CD34+ peak may be biologically primed for efficient recovery even before the final graft dose is known. A patient with a weak mobilization peak may deserve closer scrutiny, even if the collected product eventually crosses a conventional adequacy threshold. The biology of release and the biology of recovery may be linked by the same marrow reserve.
This does not replace the infused dose. The study instead suggests a layered model: the harvest dose remains necessary, but the mobilisation peak provides an additional readout of how readily the patient’s hematopoietic system responded to stimulation.
Joint Stratification Identified Higher-Risk Subgroups
Investigators also examined combined risk grouping by mobilisation peak and infused CD34+ dose. Patients were stratified into subgroups that separated into different engraftment-risk profiles, and the combined approach identified groups at high risk of delayed neutrophil recovery.
This combined framework is probably where the study is most clinically useful. An individual with a seemingly adequate infused dose but a low mobilisation peak may not behave like a patient with both a strong peak and a high collected dose. In transplant medicine, the difference between adequate and robust can matter when the marrow is stressed by high-dose chemotherapy and the patient is waiting for neutrophil recovery to begin.
What This Means for Day-to-Day Transplant Practice
The cohort was not designed to replace standard CD34+ dose thresholds. It was designed to ask whether a biologic signal available before collection could sharpen prediction of engraftment timing. The answer was yes.
For transplant teams, the practical implication is straightforward: peripheral blood CD34+ peak during mobilisation deserves attention as a predictive variable, especially when planning harvest timing and anticipating post-transplant recovery. It may help identify patients who are likely to engraft more slowly despite receiving what appears to be an acceptable graft dose.
Limits of the Evidence and the Remaining Uncertainty
The study was prospective but observational, and it came from a single transplant program. Mobilisation strategies were not uniform, and post-transplant G-CSF use differed by clinical unit, introducing a real-world heterogeneity that may affect recovery time. The report also relied on practice patterns from one center, including the timing of monitoring and thresholds used to trigger apheresis.
The AUC of 0.695 indicates useful but incomplete discrimination. Mobilisation peak should be treated as an adjunctive marker, not a standalone decision rule. The work does not establish a universal cutoff for clinical action, and it does not prove that changing harvest timing based on the peak will improve outcomes. What it does show is narrower and more actionable: in autologous HSCT, the peripheral blood CD34+ peak during mobilisation tracks with neutrophil engraftment even when infused dose is already adequate.