In this issue of Blood, Kordasti et al report findings on deep immunophenotyping which revealed a distinct population of regulatory T cells (Tregs) that associated with disease onset and response to immunosuppressive therapy in aplastic anemia (AA). This population, named Treg subset B, showed greater immunoregulatory properties compared with Treg subset A which were identified using robust deep-phenotyping techniques. This observation adds to the wealth of data which shows an altered immune system in AA.1 

The CyTOF density plots (visual stochastic neighbor embedding [viSNE]) revealed 2 subpopulations within Tregs, namely A and B. Their frequencies were distinct between healthy donors and patients with AA. Furthermore, patients with a higher Treg A frequency were less likely to respond to immunosuppression, and conversely, those with a higher Treg B frequency more likely to have a hematologic recovery following therapy. See Figure 2D in the article by Kordasti et al that begins on page 1193.

The CyTOF density plots (visual stochastic neighbor embedding [viSNE]) revealed 2 subpopulations within Tregs, namely A and B. Their frequencies were distinct between healthy donors and patients with AA. Furthermore, patients with a higher Treg A frequency were less likely to respond to immunosuppression, and conversely, those with a higher Treg B frequency more likely to have a hematologic recovery following therapy. See Figure 2D in the article by Kordasti et al that begins on page 1193.

An immunologic disarray forms the basis of bone marrow destruction in severe AA resulting in profound cytopenias and substantial morbidity/mortality if marrow function is not improved. Over the years, several methodologies have been applied to characterize this perturbation that culminates in marrow failure. In the early days of flow cytometry, Zoumbos et al showed an increased proportion of interferon-γ (IFNγ)-producing T cells in the blood and bone marrow of patients with AA.2  Subsequently, IFNγ and other cytokines such as tumor necrosis factor β (TNFβ) were implicated in hematopoietic suppression by upregulation of Fas in marrow progenitor cells leading to apoptosis.3  Later, advances in flow cytometric protocols allowed for the characterization of skewing of the T-cell repertoire in AA and patient-specific clonotypes were identified by T-cell receptor sequencing which tracked with disease activity.4  More broadly, the Treg subset was shown to be decreased in AA whereas the more proinflammatory TH17 subset and TH1-associated proteins increased at diagnosis.5 

The refinement of these techniques has allowed greater depth in molecular and phenotypic characterization of the genetic, molecular, and T-cell subset alterations in severe AA (SAA). Next-generation sequencing has been applied to hundreds of patients with SAA with important insights into the extent of clonal hematopoiesis and its implication in outcomes.6,7  A deeper immune subset analysis is now possible with single-cell mass cytometry (cytometry by time-of-flight [CyTOF]) which, instead of fluorescence-based flow cytometry, applies transition element isotopes to label antibodies which are analyzed by a time-of-flight mass spectrometer.8  The tagging of antibodies to rare heavy metals in CyTOF allows for the analysis of dozens of parameters simultaneously providing multidimensional data with little to no background (given the minimal autofluorescence) and no need for compensation.

Kordasti et al applied CyTOF in a 31-patient cohort providing an in-depth characterization of the Treg subset in AA. Two distinct Treg subsets named A and B were defined based on an extended phenotypic profile (12 parameters), gene expression, expandability, and suppression function. Treg B was significantly lower at diagnosis compared to healthy donors and tended to improve among responders to immunosuppression. Furthermore, subset B had more IFNγ- and TNFα-suppressing properties than subset A. There was minimal clonotypic overlap between the 2 Treg subsets, suggesting distinct origins. A higher Treg B population at baseline correlated with hematologic recovery following immunosuppression with this subpopulation increase likely contributing to regulation of autoimmunity in AA in recovering patients (see figure) although the number of patients in this correlative clinical analysis was small.

The more in-depth characterization of molecular and immunologic disarrangements in SAA has created a challenge in applying the wealth of information into the clinic. One of the more immediate goals with this data is to predict short- (hematologic response) and long-term outcomes (relapse and clonal evolution) following immunosuppressive therapy. Prognostication has long been sought in all fields of medicine as evidenced by the thousands of publications on the topic.9  However, very few change practice. Many are not applied for not being practical, reproducible, robust, or cost-effective.9  Although certain genetic defects (mutations in ASXL1, DNMT3A) and immunologic findings such as expansion of oligoclones can correlate with outcomes, these observations are highly variable, suggesting a more intricate relationship between these findings and outcomes. A more functional assessment of these findings, in the context of immunosuppression failure for example, might be more revealing. Thus, discrimination of Treg subsets with CyTOF is not likely to be applied routinely in the clinic as a prognosticator. However, this important observation continues to strengthen the wealth of data detailing the aberrant immune system in AA. Finally, recent reports with a 3-drug immunosuppression regimen which included eltrombopag along with horse antithymocyte globulin plus cyclosporine have shown very high overall hematologic response rates (around 90%) with excellent survival to date, making the prediction of response less critical.10  Predicting durability of response and the occurrence of late events such as clonal evolution with these more sophisticated tools will become more pressing as this information could have an impact on treatment decisions.

Conflict-of-interest disclosure: The author declares no competing financial interests.

REFERENCES

1
Kordasti
 
S
Costantini
 
B
Seidl
 
T
, et al. 
Deep phenotyping of Tregs identifies an immune signature for idiopathic aplastic anemia and predicts response to treatment.
Blood
2016
 
128(9):1193-1205
2
Zoumbos
 
NC
Gascón
 
P
Djeu
 
JY
Trost
 
SR
Young
 
NS
Circulating activated suppressor T lymphocytes in aplastic anemia.
N Engl J Med
1985
, vol. 
312
 
5
(pg. 
257
-
265
)
3
Maciejewski
 
JP
Selleri
 
C
Sato
 
T
Anderson
 
S
Young
 
NS
Increased expression of Fas antigen on bone marrow CD34+ cells of patients with aplastic anaemia.
Br J Haematol
1995
, vol. 
91
 
1
(pg. 
245
-
252
)
4
Risitano
 
AM
Maciejewski
 
JP
Green
 
S
Plasilova
 
M
Zeng
 
W
Young
 
NS
In-vivo dominant immune responses in aplastic anaemia: molecular tracking of putatively pathogenetic T-cell clones by TCR β-CDR3 sequencing.
Lancet
2004
, vol. 
364
 
9431
(pg. 
355
-
364
)
5
de Latour
 
RP
Visconte
 
V
Takaku
 
T
, et al. 
Th17 immune responses contribute to the pathophysiology of aplastic anemia.
Blood
2010
, vol. 
116
 
20
(pg. 
4175
-
4184
)
6
Yoshizato
 
T
Dumitriu
 
B
Hosokawa
 
K
, et al. 
Somatic mutations and clonal hematopoiesis in aplastic anemia.
N Engl J Med
2015
, vol. 
373
 
1
(pg. 
35
-
47
)
7
Kulasekararaj
 
AG
Jiang
 
J
Smith
 
AE
, et al. 
Somatic mutations identify a subgroup of aplastic anemia patients who progress to myelodysplastic syndrome.
Blood
2014
, vol. 
124
 
17
(pg. 
2698
-
2704
)
8
Kay
 
AW
Strauss-Albee
 
DM
Blish
 
CA
Application of mass cytometry (CyTOF) for functional and phenotypic analysis of natural killer cells.
Methods Mol Biol
2016
, vol. 
1441
 (pg. 
13
-
26
)
9
Ioannidis
 
JP
Tzoulaki
 
I
What makes a good predictor? The evidence applied to coronary artery calcium score.
JAMA
2010
, vol. 
303
 
16
(pg. 
1646
-
1647
)
10
Townsley
 
DM
Dumitriu
 
B
Scheinberg
 
P
, et al. 
Eltrombopag added to standard immunosuppression for aplastic anemia accelerates count recovery and increases response rates [abstract].
Blood
2015
 
126(23). Abstract LBA-2