BACKGROUND: Increasingly large and complex biomedical data sets challenge conventional hypothesis-driven analytical approaches, however, data-driven unsupervised learning can detect inherent patterns in such data sets. METHODS: While unsupervised analysis in the medical literature commonly only utilizes a single clustering algorithm for a given data set, we developed a large-scale model with 605 different combinations of target dimensionalities as well as transformation and clustering algorithms and subsequent meta-clustering of individual results. With this model, we investigated a large cohort of 1383 patients from 59 centers in Germany with newly diagnosed acute myeloid leukemia for whom 212 clinical, laboratory, cytogenetic and molecular genetic parameters were available. RESULTS: Unsupervised learning identifies four distinct patient clusters, and statistical analysis shows significant differences in rate of complete remissions, event-free, relapse-free and overall survival between the four clusters. In comparison to the standard-of-care hypothesis-driven European Leukemia Net (ELN2017) risk stratification model, we find all three ELN2017 risk categories being represented in all four clusters in varying proportions indicating unappreciated complexity of AML biology in current established risk stratification models. Further, by using assigned clusters as labels we subsequently train a supervised model to validate cluster assignments on a large external multicenter cohort of 664 intensively treated AML patients. CONCLUSIONS: Dynamic data-driven models are likely more suitable for risk stratification in the context of increasingly complex medical data than rigid hypothesis-driven models to allow for a more personalized treatment allocation and gain novel insights into disease biology.
- Publikační typ
- časopisecké články MeSH
HLA-haploidentical allogeneic hematopoietic stem cell transplantation (Haplo-HCT) is frequently used as treatment for patients with active acute myeloid leukemia (AML). Here, we investigated whether 9/10 HLA-mismatched unrelated donor transplantation (MMUD-HCT) with post-transplant cyclophosphamide (PTCy) is an adequate alternative. Inclusion criteria in this retrospective registry study consisted of adult patients, first HCT with a Haplo donor or MMUD between 2010 and 2020 using PTCy as graft-versus-host disease (GVHD) prophylaxis, and primary refractory or relapsed disease. MMUD patients were pair-matched 1 to 2 with Haplo-recipients. A total of 73 MMUD patients met the inclusion criteria. Their data were compared to those of 146 Haplo patients in a matched-pair analysis. Median follow-up was 27 months in MMUD patients and 36 months in Haplo recipients. Two-year incidences of relapse and non-relapse mortality (NRM) were 40% and 18% in MMUD patients, respectively, versus 50% (P = 0.23) and 24% (P = 0.18) in Haplo recipients. Two-year leukemia-free survival (LFS) and overall survival (OS) was 42% and 46% in MMUD recipients, respectively, versus 26% (P = 0.1) and 28% (P = 0.061) in Haplo-patients. In conclusions, in AML patients with active disease at transplantation, MMUD-HCT results in at least comparable outcomes to Haplo-HCT when PTCy is applied.
- MeSH
- akutní myeloidní leukemie * farmakoterapie MeSH
- cyklofosfamid terapeutické užití MeSH
- dospělí MeSH
- haploidentická transplantace škodlivé účinky MeSH
- lidé MeSH
- nemoc štěpu proti hostiteli * etiologie MeSH
- nepříbuzný dárce MeSH
- příprava pacienta k transplantaci metody MeSH
- retrospektivní studie MeSH
- transplantace hematopoetických kmenových buněk * metody MeSH
- Check Tag
- dospělí MeSH
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH
Allogeneic hematopoietic cell transplantation (HCT) is recommended in high-risk patients with T cell acute lymphoblastic leukemia (T-ALL). For patients without an HLA-identical donor, haploidentical (haplo-) HCT is becoming the leading source of stem cell donation. However, data are scarce on predictive factors for outcome in that setting. We identified 122 adults (20% female; median age, 31 years; range, 18 to 68 years) with T-ALL who underwent haplo-HCT with post-transplantation cyclophosphamide (ptCy) between 2010 and 2017. The median duration of follow-up of living patients was 23 months. The 2-year incidences of relapse and nonrelapse mortality were 45% and 21%, respectively. The 2-year leukemia-free survival (LFS), overall survival (OS), and graft-versus-host disease, relapse-free survival (GRFS) were 34%, 42%, and 27%, respectively. The 2-year LFS and OS were highly influenced by disease status at transplantation, being 49% and 55%, respectively, for patients in first complete remission (CR1); 34% and 50%, respectively, for those in second CR (CR2); and 8% and 12%, respectively, for patients with active disease. On multivariate analysis, only disease status was found to affect LFS and OS. Transplantation in CR2 negatively affected LFS, whereas active disease at the time of haplo-HCT negatively affected LFS and OS. In conclusion, haplo-HCT with ptCy produced encouraging results in this challenging disease, particularly when performed in patients in CR. Despite the limitation of the small sample size, our results were not affected by the type of conditioning, calling into question the need for total body irradiation-based myeloablative conditioning in that setting.
- MeSH
- akutní myeloidní leukemie * MeSH
- cyklofosfamid terapeutické užití MeSH
- dospělí MeSH
- haploidentická transplantace MeSH
- kostní dřeň MeSH
- lidé středního věku MeSH
- lidé MeSH
- lymfoblastická leukemie-lymfom z prekurzorových T-buněk * MeSH
- mladiství MeSH
- mladý dospělý MeSH
- nemoc štěpu proti hostiteli * prevence a kontrola MeSH
- přežití po terapii bez příznaků nemoci MeSH
- příprava pacienta k transplantaci MeSH
- retrospektivní studie MeSH
- senioři MeSH
- T-lymfocyty MeSH
- transplantace hematopoetických kmenových buněk * MeSH
- Check Tag
- dospělí MeSH
- lidé středního věku MeSH
- lidé MeSH
- mladiství MeSH
- mladý dospělý MeSH
- mužské pohlaví MeSH
- senioři MeSH
- ženské pohlaví MeSH
- Publikační typ
- časopisecké články MeSH
