Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Anti-thymocyte globulin’s activity against acute myeloid leukemia stem cells


Rabbit anti-thymocyte globulin (ATG (Thymoglobulin)) kills T cells in vitro and probably also in vivo as it prevents graft-vs-host disease (GvHD) in patients. Recently we demonstrated that ATG at a clinically relevant concentration (10–50 mg/L) kills in vitro not only T cells but also leukemic blasts. In the present study, we investigated whether ATG kills not only leukemic blasts but also leukemic stem cells (LSCs). We used a flow cytometric assay of complement-mediated cytotoxicity (CDC). ATG-induced death of acute myeloid leukemia (AML) cells from patients newly diagnosed with AML was measured among blasts as well as LSCs. At 10 mg/L ATG, blasts but not LSCs were killed. At 50 mg/L ATG, both blasts and LSCs were killed. We also measured ATG-mediated killing of healthy individuals’ hematopoietic stem cells (HSCs). Median 2% HSCs from blood and 15% HSCs from filgrastim-mobilized grafts were killed with 50 mg/L ATG, compared to 30% LSCs from the blood of AML patients (p = 0.001 and 0.022, respectively). In conclusion, LSCs are sensitive to ATG, however, only at a relatively high ATG concentration. At that concentration, LSCs are killed to a higher degree than HSCs.


Allogeneic hematopoietic cell transplantation (HCT) is a potential cure for patients with acute myeloid leukemia (AML). However, leukemia relapse and graft-versus-host disease (GvHD) are two major limitations to the success of HCT [1]. Small molecule immunosuppressive drugs such as cyclosporine and methotrexate, although effective for decreasing the incidence of GvHD, increase the incidence of relapse [2, 3]. Interestingly, polyclonal rabbit anti-human thymocyte globulin (ATG, Thymoglobulin) added to myeloablative conditioning has been shown to reduce GvHD without affecting relapse [4,5,6]. The reason why ATG decreases GvHD but does not increase relapse is unclear. A potential explanation could be that ATG kills not only donor-anti-host T cells, responsible for mediating GvHD and graft-versus-leukemia effect (GVL), but also leukemic cells [7,8,9]. Recently, we showed that ATG induces complement-dependent cytotoxicity (CDC) and complement-independent cytotoxicity (CIC) of leukemic blasts [7]. The killing of the blasts was relatively minor at 10 mg/L, which is the median maximum ATG serum concentration achieved in patients (range 1–25 mg/L) who received 4.5 mg/kg ATG [7]. Major killing of the blasts was achieved with higher concentrations [7], including 50 mg/L, which is the median maximum ATG serum concentration achieved in patients who received 6 mg/kg ATG [10]. There is a high patient-to-patient variability in the sensitivity of blasts to ATG—most are sensitive, but some appear resistant to the ATG-mediated killing [7].

It is unclear whether the “no increase in relapse” applies only to Thymoglobulin or also to Grafalon (rabbit anti-Jurkat T cell globulin). The “no increase in relapse” has been documented for Thymoglobulin in 3 of 3 randomized studies [4,5,6]. For Grafalon, there have been four randomized studies; two of them showed no increase in relapse [11, 12], but the remaining two studies showed a trend toward increased incidence of relapse [13, 14]. Our present study is focused on Thymoglobulin. Throughout this manuscript, the abbreviation “ATG” refers to anti-thymocyte globulin (Thymoglobulin) unless otherwise stated.

Relapse is the primary cause of mortality after chemotherapy without as well as with HCT [15,16,17,18]. The reason for relapse remains unclear. However, it has been long proposed that relapse occurs due to the persistence of a rare subset of relatively quiescent malignant cells that are not eradicated efficiently with current treatments, the so-called leukemic stem cells (LSCs) [19,20,21]. LSCs are defined as cells that (1) are capable of initiating the disease in an immunodeficient animal as well as sustaining the disease through multiple serial transplants (self-renewal) [22, 23], and (2) give rise to the bulk of leukemic cells. LSCs are difficult to eradicate with standard chemotherapy. This leads to disease resistance and relapse. It is thought that active anti-LSC therapies would prevent relapse. It is not known whether ATG kills LSCs.

AML LSCs are known to be more similar to healthy hematopoietic stem cells (HSCs) than to their differentiated progeny, including CD34+ CD38 phenotype and capacity to self-renew [22,23,24]. Therefore, ideal anti-LSC therapy must spare normal HSCs to preserve long-term hematopoiesis or prevent graft failure after HCT.

Here we studied the effect of ATG on leukemic blasts, LSCs, and HSCs.

Patients and methods

Patients and donors

To study LSCs, we obtained blood samples from 27 patients newly diagnosed with AML (before induction chemotherapy). Patient characteristics are shown in Table 1. To study healthy HSCs, we obtained blood from 10 healthy volunteers and filgrastim-mobilized blood stem cells (graft specimens) from 17 allogeneic graft donors. Characteristics of the healthy volunteers and the graft donors are shown in Table 1. Informed consents were obtained from both the patients and the donors. The study was approved by the Health Research Ethics Board of Alberta.

Table 1 Patient and donor characteristics

Determination of ATG cytotoxicity against stem cells by flow cytometry

Despite some controversy, both LSCs and HSCs are believed to be enriched in the CD34+ CD38 compartments [25]. Here we assumed that the CD34+ CD38 cells from AML patients at diagnosis contain virtually only LSCs (the number of healthy HSCs being negligible) and that the CD34+ CD38 cells from healthy volunteers contain only healthy HSCs (and no LSCs). Thus, both LSCs and HSCs were defined as CD45dim/–, side scatter low (SSclow), lineage (not expressing CD3, CD19, CD14, CD56, CD16, CD41a, or CD235a), CD34+, CD38 cells (Fig. 1). Blasts were defined as CD45dim/–, SSclow, and lineage cells (Fig. 1). The borders between CD38 positive and negative cells and CD34 positive and negative cells were determined using fluorescence minus one (FMO) controls (Supplementary Fig. 1).

Fig. 1

Gating strategy to identify leukemic stem cells (LSCs) and healthy hematopoietic stem cells (HSCs). LSCs and HSCs were defined as mononuclear cells that were CD45dim/–, SSclow, did not express lineage antigens (CD19, CD14, CD16, CD56, CD3, CD235a, and CD41a), expressed CD34 and did not express CD38. CD38 fluorescence minus one (FMO) and CD34 FMO were used to correctly gate the LSC and HSC population (Supplementary Fig. 1). Doublets were discriminated using FSc-W vs FSc-H and SSc-W vs SSc-H plots (not shown). FSc forward scatter, SSc side scatter, FSc-W forward scatter-width, FSc-H forward scatter-height, SSc-W side scatter-width, SSc-H side scatter-height

Percent dead cells (7-aminoactinomycin D (7AAD) positive) after exposure to ATG was measured in CDC assay. Percent apoptotic cells (Annexin V positive) after exposure to ATG was measured in CIC assay. The details of these assays, as well as the statistical analysis are described in Supplementary Appendix.


ATG at 50 mg/L, but not 10 mg/L, kills leukemic stem cells

CDC of LSCs was induced by 50 mg/L but not at 10 mg/L ATG (Fig. 2). At the 50 mg/L concentration, the median-adjusted (background subtracted) percent 7AAD+ LSCs was 38%. The sensitivity of LSCs to ATG at the 50 mg/L concentration was highly variable—the adjusted percent of 7AAD+ LSCs ranged from 0.2 to 98% (Supplementary Fig. 2).

Fig. 2

ATG induces complement-dependent cytotoxicity (CDC) of leukemic stem cells (LSCs). Top: Histograms gated on LSCs from a typical patient, exposed to 10 mg/L and 50 mg/L of ATG in the presence of complement (red) or the absence of complement (gray) for 15 minutes. ATG concentration of 300 mg/L was used as a positive control—percent 7AAD+ mononuclear cells (not LSCs) was determined. Values in the histograms represent percents of 7AAD+ LSCs. Bottom: Aggregate results in 20 AML patients. The median and the range are shown for each condition. LSCs are defined as mononuclear cells that were CD45dim/–, SSclow, Lineage (CD19, CD14, CD16, CD56, CD3, CD235a, and CD41a), CD34+, and CD38. LSC leukemia stem cell, 7AAD 7-aminoactinomycin d (a marker of dead cells)

Similar to CDC, CIC of LSCs was induced by 50 mg/L but not 10 mg/L ATG (Fig. 3). Contrary to CDC, even at the 50 mg/L concentration, the ATG-mediated CIC appeared weak—the median-adjusted percent Annexin V+ LSCs was only 13%. Similar to CDC, the sensitivity of LSCs to 50 mg/L ATG was variable—the adjusted percent Annexin V+ LSCs ranged from 0 to 37% (Supplementary Fig. 3).

Fig. 3

ATG-induced complement-independent cytotoxicity (CIC) of leukemic stem cells (LSCs). Top: Dot plots gated on LSCs from a typical patient, exposed to 10 mg/L and 50 mg/L ATG for 4 h. ATG concentration of 300 mg/L and ultraviolet light (UV) were used as positive controls—percent Annexin V+ mononuclear cells (not LSCs) was determined. Values in the dot plots represent percentages of Annexin V+ LSCs. Bottom: Aggregate results in 20 AML patients. The median and the range are shown for each condition. LSCs are defined as mononuclear cells that were CD45dim/–, SSclow, Lineage (CD19, CD14, CD16, CD56, CD3, CD235a, and CD41a), CD34+, and CD38. LSC leukemia stem cell, Annexin V a marker of apoptotic cells

For comparison, the killing of T cells in the patients was as follows: For CDC, the median-adjusted percent 7AAD+ T cells was 15% at 10 mg/L ATG (vs 0.5% for LSCs) and 59% at 50 mg/L ATG (vs 38% for LSCs) (Supplementary Fig. 4). For CIC, the median-adjusted percent Annexin V+ T cells was 17% at 10 mg/L (vs 7% for LSCs) and 27% at 50 mg/L (vs 13% for LSCs) (Supplementary Fig. 5).

Leukemic blasts are more sensitive to ATG than LSCs via CDC, but equally sensitive via CIC

Due to phenotypic and functional differences between leukemic blasts and LSCs [26, 27], ATG might have a different effect on leukemic blasts versus LSCs. Therefore, we compared leukemic blasts and LSCs in ATG-mediated CDC and CIC. We did this in 20 AML patients.

In the CDC assay, leukemic blasts were killed to a greater degree than LSCs (Fig. 4). At 10 mg/L ATG, the median-adjusted percent 7AAD+ cells was 3% among blasts vs 0.5% among LSCs (p = 0.006). At 50 mg/L ATG, it was 64% among blasts vs 38% among LSCs (p = 0.009).

Fig. 4

Leukemic blasts are more sensitive to ATG-mediated killing via CDC than LSCs. Graphs show the adjusted (background subtracted) percentage of 7AAD+ cells among LSCs (blue) and leukemic blasts (red). Cells were incubated with 10 mg/L (left) or 50 mg/L (right) ATG. Medians are shown as horizontal lines. The significance of the difference of the adjusted percentages of 7AAD+ blasts vs LSCs was determined using Wilcoxon signed-rank test. LSC leukemia stem cell, 7AAD 7-aminoactinomycin d (a marker of dead cells)

In the CIC assay, at both 10 and 50 mg/L ATG, the killing of both blasts and LSCs was minor. In contrast to CDC, at both 10 and 50 mg/L concentration, there was no difference in killing of leukemic blasts vs LSCs (Supplementary Fig. 6).

Complement is not consumed in patients receiving standard dose of ATG

Given that LSCs were observed to be killed by ATG in the CDC assay at 50 mg/L but not 10 mg/L ATG, it might be theoretically desirable to use a high ATG dose (resulting in serum levels of ≥50 mg/L, achieved after 6–8 mg/kg ATG) [10], as this could minimize both relapse and GvHD. However, it is not known whether complement is depleted in patients after they have received a lower ATG dose. To address this question, we performed CDC assay against human lymphocytes (obtained from a healthy volunteer). We used serum of patients who received our conventional dose of ATG (0.5 mg/kg on day −2, 2.0 mg/kg on day −1, and 2.0 mg/kg on day 0, expected to lead to pre-graft infusion ATG concentration of ~10 mg/L) [7] as the source of complement. The sera were obtained from 10 allogeneic HCT patients on day 0, between finishing ATG infusion and starting graft infusion. The allogeneic HCT patient sera were compared to sera of autologous HCT patients, who did not receive ATG, in their ability to lyse lymphocytes with ATG at 50 mg/L and 300 mg/L. Pooled serum from healthy volunteers was also used for comparison. As negative controls, we used cells alone (without serum and without ATG), cells incubated with serum but no ATG, and cells incubated with ATG in the presence of heat-inactivated sera (56 °C for 1 h to destroy complement). As shown in Fig. 5, percent dead (7AAD+) lymphocytes after 15 min incubation with 50 mg/L ATG was higher when using sera from patients who received ATG (allogeneic HCT patients) vs patients who did not receive ATG (autologous HCT patients) (median 79% vs 72%, p = 0.03). Similar result was obtained for ATG at 300 mg/L. This suggests that complement is not depleted after administration of 4.5 mg/kg ATG. On the contrary, there appears to be more complement in the ATG-treated patients than the non-ATG-treated patients or healthy volunteers. The reason is unclear; it is conceivable that the partial consumption of complement on day −2 and −1 stimulated the liver to release or produce more complement. We also confirmed the “no depletion of complement” in an alternative experiment where we performed the CDC assay against HPB-ALL cells (T-cell line) instead of primary lymphocytes. As shown in Supplementary Fig. 7, percent dead (7AAD+) HPB-ALL cells after 15 min incubation with 300 mg/L ATG was similar when using sera from patients vs healthy volunteers (median 98% vs 96%, respectively). Thus, complement is not depleted after the standard ATG dose of 4.5 mg/kg.

Fig. 5

Complement is not consumed in patients who received ATG versus who did not. Sera from 10 patients undergoing allo-HCT who received the standard dose of ATG (4.5 mg/kg) (red), from 10 patients undergoing auto-HCT who did not receive any ATG (green) and from healthy volunteers (pooled) (green) were used. MNCs from a healthy volunteer were incubated with the serum samples with and without ATG. Percentage of dead (7AAD+) cells is shown on the y-axis. Serum without complement means heat-inactivated serum. The bars denote median and range. A.serum active serum (complement not destroyed by heat inactivation), HI heat-inactivated, HV healthy volunteers, Auto autologous, Allo allogeneic, HCT hematopoietic cell transplantation, 7AAD 7-aminoactinomycin d (a marker of dead cells)

ATG kills LSCs to a greater degree than healthy HSCs

Given similarities between LSCs and healthy HSCs [22, 28], the effect of ATG on healthy HSCs should be considered. We compared ATG-mediated killing of LSCs from the blood of 10 AML patients vs HSCs from the blood of 10 healthy volunteers. As HSCs in healthy blood are sparse (Table 1), we immunomagnetically enriched CD34+ cells from the blood of both the patients and the volunteers. Given limited number of CD34+ cells obtained from the volunteers, we performed only the CDC (and not CIC) assay, since CDC appears to be the more important mechanism of ATG-induced killing (compare Figs. 2 and 3), and used only the 50 mg/L ATG concentration, which is the relevant concentration for killing LSCs (Fig. 2). As shown in Fig. 6, the median percents dead (7AAD+) LSCs and HSCs after treatment with 50 mg/L ATG were 40% (range 1 to 90%) and 1% (range 0 to 30%), respectively (p = 0.001). This suggests that LSCs are more sensitive to ATG-induced killing than HSCs.

Fig. 6

LSCs are more sensitive to ATG-mediated killing via CDC than healthy HSCs. Data are shown on 10 AML patients whose blood LSCs (blue) we studied, and on 10 healthy volunteers whose blood HSCs (red) we studied (side-by-side, i.e., in one experiment, one patient, and one healthy volunteer were studied). The MNCs from AML patients as well as from the healthy volunteers were first enriched for CD34+ cells using immunomagnetic beads followed by treatment with ATG and finally staining for Annexin V to measure CDC of LSCs and HSCs. Adjusted percentage (background subtracted) is shown for each patient/healthy volunteer. Medians are shown as horizontal lines. Significance of the difference in the percentage of 7AAD+ LSCs vs HSCs was determined using Mann–Whitney U test. LSC leukemia stem cell, HSCs hematopoietic stem cells, 7AAD 7-aminoactinomycin d (a marker of dead cells)

During HCT, patients receive apheresed MNCs from donors treated with filgrastim (to mobilize HSCs from marrow to blood). Thus, comparing LSCs to HSCs from the grafts (post-filgrastim) may be clinically more relevant than to HSCs from the blood of healthy volunteers (who have not received filgrastim). Therefore, we compared LSCs from the blood of 17 AML patients to HSCs from 17 graft samples. No immunomagnetic enrichment for CD34+ cells was done (as percentages of CD34+ cells in both patient blood and donor grafts were sufficient). As shown in Fig. 7, median percents 7AAD+ LSCs and graft HSCs after treatment with 50 mg/L ATG were 39% (range 0 to 98%) and 16% (range 0 to 39%), respectively (p = 0.022). This suggests that LSCs are also more sensitive to ATG-induced complement-mediated lysis than graft HSCs.

Fig. 7

LSCs are more sensitive to ATG-mediated CDC than filgrastim-mobilized HSCs. Data are shown on 17 AML patients whose LSCs (blue) we studied and on grafts from 17 hematopoietic cell donors (green) we studied (side-by-side, i.e., in one experiment, one patient, and one graft were studied). Adjusted percentage (background subtracted) is shown for each patient/donor. Medians are shown as horizontal lines. Significance of the difference in the percentage of 7AAD+ LSCs vs graft-HSCs was determined using Mann–Whitney U test. Medians are shown as horizontal lines. LSC leukemia stem cell, Graft HSCs hematopoietic stem cells (derived from filgrastim-mobilized grafts), 7AAD 7-aminoactinomycin d (a marker of dead cells)

Unlike the case of healthy HSCs from blood, in the graft specimens, we were able to perform not only CDC but also CIC assays due to sufficient number of HSCs. Thus, we evaluated CIC (using 50 mg/L ATG) against LSCs and graft-HSCs. Similar to CDC, the LSCs were significantly more sensitive to ATG in the CIC assay than the graft HSCs (p = 0.036) (Supplementary Fig. 8).

Anti-leukemic activity of Thymoglobulin versus Grafalon

We wished to compare Thymoglobulin and Grafalon in their ability to kill (by CDC) LSCs, including at concentrations above 50 mg/L as Grafalon is known to kill immune cells at higher concentrations than Thymoglobulin [29, 30]. However, our CDC assay did not provide meaningful results when >50 mg/L ATG (Thymoglobulin or Grafalon) was used. At those concentrations ATG outcompeted fluorochrome-labeled antibodies in binding to the antigens used to define LSCs. Therefore, we decided to instead compare Grafalon and Thymoglobulin in their ability to kill leukemic cells (blasts + LSCs). We used blood mononuclear cells (MNCs) from 10 AML patients with very high blast percentage (median 90% CD45dim/– CD34+ cells among the MNCs, range 79–96%) and assumed that virtually all the MNCs were leukemic cells. This avoided the need to identify cells by surface antigens. As shown in Supplementary Fig. 9, both Grafalon and Thymoglobulin killed leukemic cells, however, a higher concentration of Grafalon was needed to achieve the same effect—approximately 40% leukemic cells were killed by 100 mg/L Grafalon or 50 mg/L Thymoglobulin. Thus, we conclude that both Grafalon and Thymoglobulin kill (by CDC) leukemic cells. However, as previously shown for immune cells [29, 30], a higher concentration of Grafalon is needed to achieve the same degree of killing of leukemic cells (by CDC). This is reminiscent of CIC, which was induced in leukemic cells from 4 of 8 AML patients using a relatively high concentration of Grafalon (250 mg/L) [31].


Previously, we demonstrated that ATG (Thymoglobulin) at clinically relevant concentration induces cytotoxicity against leukemic blasts [7]. In the present study, we show that ATG induces cytotoxicity also against LSCs. However, the cytotoxicity occurred with 50 mg/L and not 10 mg/L, the latter being the approximate median maximum concentration achieved in patients at our center using 4.5 mg/kg ATG. Nonetheless, 50 mg/L is clinically achievable—Remberger et al. reported median maximum concentration ≥50 mg/L using 6.0 mg/kg ATG and ≥80 mg/L using 8.0 mg/kg ATG [10]. As even higher ATG doses (10–40 mg/kg) have been used in clinic [32,33,34], it is likely that even higher ATG concentrations (>50 mg/L) can be achieved in patients.

Chemotherapeutic agents clinically used today eradicate blasts/rapidly dividing cells. However, the same agents demonstrate little to no effect on the blast progenitor cells, i.e., the LSCs [35, 36]. For example, cytarabine has virtually no activity against LSCs [37]. However when tested on leukemic blasts from the same patients, cytarabine shows potent cytotoxicity against blasts. Here we showed that Thymoglobulin kills both blasts and LSCs, albeit LSCs only at a relatively high concentration.

Even though the majority of the studies indicate that phenotypically LSCs resemble more HSCs than their mature progeny, recent studies on AML suggest that some differences between LSCs and HSC exist [38,39,40]. Therefore, it is possible that LSCs might be more or less sensitive to therapy than HSCs. From a therapeutic perspective, it is crucial, when studying the anti-leukemic activity of drugs against LSCs, to also consider their effect against HSCs to account for non-specific toxicities such as graft failure or delayed engraftment. Both Thymoglobulin and Grafalon are known to delay engraftment [4, 41]. Whereas Thymoglobulin has not been conclusively shown to induce graft failure, one study has shown that the incidence of graft failure was higher in patients treated with Grafalon compared to placebo [14]. Our results suggest that via CDC, neither LSCs nor HSCs are sensitive to ATG at 10 mg/L. With a fivefold higher concentration, both LSCs and HSCs are sensitive to ATG. However, LSCs are more sensitive. The reason for the different sensitivity of LSCs vs HSCs is unclear. Perhaps, LSCs express antigens targetted by ATG not expressed or less expressed by HSCs. For example, ATG is known to contain antibodies against CD44, and the expression of CD44 is higher on LSCs than HSCs [9, 42, 43]. Moreover, treatment with H90, an anti-CD44 antibody was shown to eradicate AML LSCs in vivo in immunodeficient mice while not altering normal hematopoiesis, making it an attractive target for LSCs therapy while sparing HSCs [42]. In addition to CD44, ATG also contains antibodies against CD25, CXCR4, CD32 which are either exclusively expressed on LSCs (and not HSCs) or more expressed on LSCs than on HSCs [9, 44].

What is the relevance of our in vitro finding of ATG killing LSCs to patients, given that so far no study has shown a lower incidence of relapse with ATG compared to no ATG and that ATG is expected to kill not only LSCs but also donor T cells mediating graft-vs-leukemia effect (GVL)? We speculate that the “no decrease in relapse” could be due to different effects of pre-HCT vs post-HCT ATG area under the curve (AUC) on relapse. Only the post-HCT AUC (but not the pre-HCT AUC) is expected to interfere with the GVL. Consistent with that, a high post-HCT AUC has been associated with a trend toward increased incidence of relapse [45]. Theoretically (if our in vitro observation applies to in vivo), a high pre-HCT AUC should be associated with a decreased incidence of relapse. Consistent with that, we have preliminary data showing a trend towards decreased incidence of relapse in patients with a high pre-HCT AUC and a trend towards increased incidence of relapse in patients with a high post-HCT AUC. If this is confirmed, it will suggest that administration of ATG designed to result in high pre-HCT AUC and relatively low post-HCT AUC (high-dose ATG given early in conditioning) could be beneficial for minimizing relapse.

There are several limitations in our study: (1) We defined both LSCs and HSCs as CD34+ and CD38, as is common practice in other studies [46]. However, some studies show that LSCs and HSCs are not limited to the CD34+ CD38 compartment but are also present in the CD34+ CD38+ or the CD34 compartment [47, 48]. Nonetheless, both in vitro and in vivo studies have provided compelling evidence that the CD34+ CD38 LSCs are probably the most relevant LSCs. For example, CD34+ CD38 frequency, but not CD34+ CD38+, total CD34+, and CD34 frequencies at diagnosis correlates with engraftment in immunocompromised mice or with relapse rates in patients [25, 49]. We could have theoretically overcome this limitation by using additional markers of LSCs, but in practice, generally accepted additional LSC markers are lacking. Following the initial publications that supported the LSC activity to be limited to the CD34+ CD38 population [22, 23], additional identification markers for LSCs have been proposed including CD123 [38, 50], CD32 [51], CD33 [52], CD45RA [53], CD47 [44], CD96 [54], CD99 [55], IL1RAP [56], and TIM-3 [57]. However, to date, there is no standard phenotype for defining LSCs in AML. Therefore, and because of the high intra-patient and inter-patient heterogeneity of AML LSC populations [47, 58,59,60], we defined LSCs as per the conventional phenotype, i.e., CD45dim/–, SSclow, Lin, CD34+, and CD38. (2) Due to the lack of a sophisticated antigen panel, we were unable to differentiate between LSCs and HSCs in our study phenotypically. Therefore, it is possible that the LSCs defined in our study could also contain healthy HSCs. However, the fraction of healthy HSCs (defined as CD34+ cells) is meager (0.01% of the total nucleated cells in blood) [61]. Given the high percentage of CD34+ CD38 cells among the leukemic blasts in the majority of the specimens (Table 1), we assume that the majority of cells having the CD45dim/–, SSclow, Lineage, CD34+, CD38 phenotype are LSCs. (3) Our experiments were performed using blood, whereas the marrow is the “home” tissue of both LSCs and HSCs. Therefore, comparison of bone marrow-derived LSCs with bone marrow-derived HSCs would have been more relevant. However, the feasibility of a comparison of marrow LSCs versus marrow HSCs is limited by practical and ethical constraints of obtaining marrow from healthy volunteers. (4) The CD34+CD38 population may contain not only LSCs or HSCs but also their progeny. We cannot exclude the possibility that the killed (7AAD+ or Annexin V+) cells among the CD34+CD38 cells belonged to the progeny rather than the LSCs/HSCs. (5) We measured the anti-leukemic activity only in vitro. Studying the anti-leukemic activity in vivo using immunocompromised mice harboring human leukemia could be more informative. However, it may also have the disadvantage of using mouse complement, which may be triggered with the rabbit ATGs to a different degree than human complement.

Despite the limitations mentioned above, our study is the first to demonstrate the effect of ATG against LSCs and, to a lesser degree, HSCs. The anti-LSC effect is, however, significant only at a relatively high ATG concentration, achievable probably with ATG doses of ≥6 mg/kg. Targeting high pre-transplant ATG AUC might lead to further reduction in GvHD [62] and possibly also relapse, if what we have described in vitro applies in vivo.


  1. 1.

    Mohty M, Apperley JF. Long-term physiological side effects after allogeneic bone marrow transplantation. Hematology Am Soc Hematol Educ Program. 2010;2010:229–36.

    Article  PubMed  Google Scholar 

  2. 2.

    Bacigalupo A, Van Lint MT, Occhini D, Gualandi F, Lamparelli T, Sogno G, et al. Increased risk of leukemia relapse with high-dose cyclosporine A after allogeneic marrow transplantation for acute leukemia. Blood. 1991;77:1423–8.

    CAS  PubMed  Google Scholar 

  3. 3.

    Storb R, Deeg J, Pepe M, Appelbaum F, Anasetti C, Beatty P, et al. Methotrexate and cyclosporine versus cyclosporine alone for prophylaxis of graft-versus-host disease in patients given HLA-identical marrow grafts for leukemia: Long-term follow-up of a controlled trial. Blood. 1989;73:1729–34.

    CAS  PubMed  Google Scholar 

  4. 4.

    Walker I, Panzarella T, Couban S, Couture F, Devins G, Elemary M, et al. Pretreatment with anti-thymocyte globulin versus no anti-thymocyte globulin in patients with haematological malignancies undergoing haemopoietic cell transplantation from unrelated donors: a randomised, controlled, open-label, phase 3, multicentre trial. Lancet Oncol. 2016;17:164–73.

    CAS  Article  PubMed  Google Scholar 

  5. 5.

    Chang YJ, Wang Y, Mo XD, Zhang XH, Xu LP, Yan CH, et al. Optimal dose of rabbit thymoglobulin in conditioning regimens for unmanipulated, haploidentical, hematopoietic stem cell transplantation: Long-term outcomes of a prospective randomized trial. Cancer. 2017;123:2881–92.

    CAS  Article  PubMed  Google Scholar 

  6. 6.

    Bacigalupo A, Lamparelli T, Barisione G, Bruzzi P, Guidi S, Alessandrino PE, et al. Thymoglobulin prevents chronic graft-versus-host disease, chronic lung dysfunction, and late transplant-related mortality: long-term follow-up of a randomized trial in patients undergoing unrelated donor transplantation. Biol Blood Marrow Transplant. 2006;12:560–5.

    Article  PubMed  Google Scholar 

  7. 7.

    Dabas R, Lee R, Servito MT, Dharmani-Khan P, Modi M, van Slyke T, et al. Antithymocyte globulin at clinically relevant concentrations kills leukemic blasts. Biol Blood Marrow Transplant. 2016;22:815–24.

    CAS  Article  PubMed  Google Scholar 

  8. 8.

    Liu H, Qin Y, Wang X, Xie K, Yang Y, Zhu J, et al. Polyclonal rabbit antithymocyte globulin induces apoptosis and has cytotoxic effects on human leukemic cells. Clin Lymphoma Myeloma Leuk. 2012;12:345–54.

    CAS  Article  PubMed  Google Scholar 

  9. 9.

    Mohty M. Mechanisms of action of antithymocyte globulin: T-cell depletion and beyond. Leukemia. 2007;21:1387–94.

    CAS  Article  PubMed  Google Scholar 

  10. 10.

    Remberger M, Persson M, Mattsson J, Gustafsson B, Uhlin M. Effects of different serum-levels of ATG after unrelated donor umbilical cord blood transplantation. Transpl Immunol. 2012;27:59–62.

    CAS  Article  PubMed  Google Scholar 

  11. 11.

    Socié G, Schmoor C, Bethge WA, Ottinger HD, Stelljes M, Zander AR et al. Chronic graft-versus-host disease: long-term results from a randomized trial on GvHD prophylaxis with or without anti–T-cell globulin ATG-fresenius. Blood. 2011.

    CAS  Article  Google Scholar 

  12. 12.

    Locatelli F, Bernardo ME, Bertaina A, Rognoni C, Comoli P, Rovelli A, et al. Efficacy of two different doses of rabbit anti-T-lymphocyte globulin to prevent graft-versus-host disease in children with haematological malignancies transplanted from an unrelated donor: a multicentre, randomised, open-label, phase 3 trial. Lancet Oncol. 2017;18:1126–36.

    CAS  Article  PubMed  Google Scholar 

  13. 13.

    Kröger N, Solano C, Wolschke C, Bandini G, Patriarca F, Pini M, et al. Antilymphocyte globulin for prevention of chronic graft-versus-host disease. New Engl J Med. 2016;374:43–53.

    CAS  Article  PubMed  Google Scholar 

  14. 14.

    Soiffer RJ, Kim HT, McGuirk J, Horwitz ME, Johnston L, Patnaik MM et al. Prospective, randomized, double-blind, Phase III Clinical Trial of anti-T-lymphocyte globulin to assess impact on chronic graft-versus-host disease-free survival in patients undergoing HLA-matched unrelated myeloablative hematopoietic cell transplantation. J Clin Oncol. 2017;35:4003–11.

    CAS  Article  Google Scholar 

  15. 15.

    Sander A, Zimmermann M, Dworzak M, Fleischhack G, Von Neuhoff C, Reinhardt D, et al. Consequent and intensified relapse therapy improved survival in pediatric AML: results of relapse treatment in 379 patients of three consecutive AML-BFM trials. Leukemia. 2010;24:1422–8.

    CAS  Article  Google Scholar 

  16. 16.

    Craddock C, Tauro S, Moss P, Grimwade D. Biology and management of relapsed acute myeloid leukaemia. Br J Haematol. 2005;129:18–34.

    CAS  Article  PubMed  Google Scholar 

  17. 17.

    Bejanyan N, Weisdorf DJ, Logan BR, Wang H-L, Devine SM, de Lima M, et al. Survival of patients with acute myeloid leukemia relapsing after allogeneic hematopoietic cell transplantation: A Center for International Blood and Marrow Transplant Research Study. Biol Blood Marrow Transplant. 2015;21:454–9.

    Article  PubMed  Google Scholar 

  18. 18.

    Lowenberg B, Griffin JD, Tallman MS Acute myeloid leukemia and acute promyelocytic leukemia. Hematology Am Soc Hematol Edu Program. 2003;2003:82–101.

    Article  Google Scholar 

  19. 19.

    van Rhenen A, Feller N, Kelder A, Westra AH, Rombouts E, Zweegman S, et al. High stem cell frequency in acute myeloid leukemia at diagnosis predicts high minimal residual disease and poor survival. Clin Cancer Res. 2005;11:6520–7.

    Article  PubMed  Google Scholar 

  20. 20.

    Costello RT, Mallet F, Gaugler B, Sainty D, Arnoulet C, Gastaut J-A, et al. Human acute myeloid leukemia CD34 progenitor cells have decreased sensitivity to chemotherapy and fas-induced apoptosis, reduced immunogenicity, and impaired dendritic cell transformation capacities. Cancer Res. 2000;60:4403–11.

    CAS  PubMed  Google Scholar 

  21. 21.

    Ishikawa F, Yoshida S, Saito Y, Hijikata A, Kitamura H, Tanaka S, et al. Chemotherapy-resistant human AML stem cells home to and engraft within the bone-marrow endosteal region. Nat Biotechnol. 2007;25:1315.

    CAS  Article  Google Scholar 

  22. 22.

    Lapidot T, Sirard C, Vormoor J, Murdoch B, Hoang T, Caceres-Cortes J, et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature. 1994;367:645–8.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  23. 23.

    Bonnet D, Dick JE. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med. 1997;3:730–7.

    CAS  Article  Google Scholar 

  24. 24.

    Bhatia M, Wang JCY, Kapp U, Bonnet D, Dick JE. Purification of primitive human hematopoietic cells capable of repopulating immune-deficient mice. Proc Natl Acad Sci USA. 1997;94:5320–5.

    CAS  Article  Google Scholar 

  25. 25.

    Terwijn M, Zeijlemaker W, Kelder A, Rutten AP, Snel AN, Scholten WJ, et al. Leukemic stem cell frequency: a strong biomarker for clinical outcome in acute myeloid leukemia. PLoS ONE. 2014;9:e107587

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  26. 26.

    Guan Y, Gerhard B, Hogge DE. Detection, isolation, and stimulation of quiescent primitive leukemic progenitor cells from patients with acute myeloid leukemia (AML). Blood. 2003;101:3142–9.

    CAS  Article  PubMed  Google Scholar 

  27. 27.

    Guzman ML, Neering SJ, Upchurch D, Grimes B, Howard DS, Rizzieri DA, et al. Nuclear factor-kappaB is constitutively activated in primitive human acute myelogenous leukemia cells. Blood. 2001;98:2301–7.

    CAS  Article  Google Scholar 

  28. 28.

    Thomas D, Majeti R. Biology and relevance of human acute myeloid leukemia stem cells. Blood. 2017.

    CAS  Article  Google Scholar 

  29. 29.

    Popow I, Leitner J, Majdic O, Kovarik JJ, Saemann MD, Zlabinger GJ, et al. Assessment of Batch to Batch Variation in Polyclonal Antithymocyte Globulin Preparations. Transplantation. 2012;93:32–40.

    CAS  Article  PubMed  Google Scholar 

  30. 30.

    Ayuk FA, Atassi N, Schuch G, Mina S, Fang L, Bokemeyer C, et al. Complement-dependent and complement-independent cytotoxicity of polyclonal antithymocyte globulins in chronic lymphocytic leukemia. Leuk Res. 2008;32:1200–6.

    CAS  Article  PubMed  Google Scholar 

  31. 31.

    Grullich C, Ziegler C, Finke J. Rabbit anti T-lymphocyte globulin induces apoptosis in peripheral blood mononuclear cell compartments and leukemia cells, while hematopoetic stem cells are apoptosis resistant. Biol Blood Marrow Transplant. 2009;15:173–82.

    Article  PubMed  Google Scholar 

  32. 32.

    Schattenberg A, van der Meer A, Preijers F, Schaap N, Rinkes M, van der Maazen R, et al. Addition of ATG to the conditioning regimen is a major determinant for outcome after transplantation with partially lymphocyte-depleted grafts from voluntary unrelated donors. Bone Marrow Transplant. 2004;33:1115–21.

    CAS  Article  PubMed  Google Scholar 

  33. 33.

    Bacigalupo A, Lamparelli T, Bruzzi P, Guidi S, Alessandrino PE, di Bartolomeo P, et al. Antithymocyte globulin for graft-versus-host disease prophylaxis in transplants from unrelated donors: 2 randomized studies from Gruppo Italiano Trapianti Midollo Osseo (GITMO). Blood. 2001;98:2942–7.

    CAS  Article  Google Scholar 

  34. 34.

    Seidel MG, Fritsch G, Matthes-Martin S, Lawitschka A, Lion T, Potschger U, et al. Antithymocyte globulin pharmacokinetics in pediatric patients after hematopoietic stem cell transplantation. J Pediatr Hematol Oncol. 2005;27:532–6.

    Article  PubMed  Google Scholar 

  35. 35.

    Kantarjian HM, Estey EH, Keating MA. New chemotherapeutic agents in acute myeloid leukemia. Leukemia. 1996;10(Suppl 1):S4–6

    PubMed  Google Scholar 

  36. 36.

    Maugeri-Saccà M, Vigneri P G, De Maria R. Cancer stem cells and chemosensitivity. Clin Cancer Res. 2011;17:4942–7.

    Article  Google Scholar 

  37. 37.

    Li TK, Houghton PJ, Desai SD, Daroui P, Liu AA, Hars ES, et al. Characterization of ARC-111 as a novel topoisomerase I-targeting anticancer drug. Cancer Res. 2003;63:8400–7.

    CAS  PubMed  Google Scholar 

  38. 38.

    Jin L, Lee EM, Ramshaw HS, Busfield SJ, Peoppl AG, Wilkinson L, et al. Monoclonal antibody-mediated targeting of CD123, IL-3 receptor α chain, eliminates human acute myeloid leukemic stem cells. Cell Stem Cell. 2009;5:31–42.

    CAS  Article  Google Scholar 

  39. 39.

    Al-Mawali A, Gillis D, Lewis I. Characteristics and prognosis of adult acute myeloid leukemia with internal tandem duplication in the FLT3gene. Oman Med J. 2013;28:432.

    CAS  Article  Google Scholar 

  40. 40.

    De Jonge HJ, Woolthuis CM, Vos AZ, Mulder A, Van Den Berg E, Kluin P, et al. Gene expression profiling in the leukemic stem cell-enriched CD34 + fraction identifies target genes that predict prognosis in normal karyotype AML. Leukemia. 2011;25:1825.

    Article  Google Scholar 

  41. 41.

    Finke J, Bethge WA, Schmoor C, Ottinger HD, Stelljes M, Zander AR, et al. Standard graft-versus-host disease prophylaxis with or without anti-T-cell globulin in haematopoietic cell transplantation from matched unrelated donors: a randomised, open-label, multicentre phase 3 trial. Lancet Oncol. 2009;10:855–64.

    CAS  Article  PubMed  Google Scholar 

  42. 42.

    Jin L, Hope KJ, Zhai Q, Smadja-Joffe F, Dick JE. Targeting of CD44 eradicates human acute myeloid leukemic stem cells. Nat Med. 2006;12:1167–74.

    CAS  Article  PubMed  Google Scholar 

  43. 43.

    Hertweck MK, Erdfelder F, Kreuzer KA. CD44 in hematological neoplasias. Ann Hematol. 2011;90:493–508.

    CAS  Article  PubMed  Google Scholar 

  44. 44.

    Majeti R, Chao MP, Alizadeh AA, Pang WW, Jaiswal S, Gibbs KD Jr., et al. CD47 is an adverse prognostic factor and therapeutic antibody target on human acute myeloid leukemia stem cells. Cell. 2009;138:286–99.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  45. 45.

    Admiraal R, Nierkens S, de Witte MA, Petersen EJ, Fleurke G-j, Verrest L, et al. Association between anti-thymocyte globulin exposure and survival outcomes in adult unrelated haemopoietic cell transplantation: a retrospective, pharmacodynamic cohort analysis. Lancet Haematol. 2017;4:e183–91.

    Article  PubMed  Google Scholar 

  46. 46.

    Kreso A, Dick JE. Evolution of the cancer stem cell model. Cell Stem Cell. 2014;14:275–91.

    CAS  Article  Google Scholar 

  47. 47.

    Taussig DC, Vargaftig J, Miraki-Moud F, Griessinger E, Sharrock K, Luke T, et al. Leukemia-initiating cells from some acute myeloid leukemia patients with mutated nucleophosmin reside in the CD34(-) fraction. Blood. 2010;115:1976–84.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  48. 48.

    Clevers H. The cancer stem cell: premises, promises and challenges. Nat Med. 2011;17:313–9.

    CAS  Article  Google Scholar 

  49. 49.

    Saito Y, Uchida N, Tanaka S, Suzuki N, Tomizawa-Murasawa M, Sone A, et al. Induction of cell cycle entry eliminates human leukemia stem cells in a mouse model of AML. Nat Biotechnol. 2010;28:275–80.

    CAS  Article  PubMed  Google Scholar 

  50. 50.

    Jordan CT, Upchurch D, Szilvassy SJ, Guzman ML, Howard DS, Pettigrew AL, et al. The interleukin-3 receptor alpha chain is a unique marker for human acute myelogenous leukemia stem cells. Leukemia. 2000;14:1777–84.

    CAS  Article  Google Scholar 

  51. 51.

    Saito Y, Kitamura H, Hijikata A, Tomizawa-Murasawa M, Tanaka S, Takagi S, et al. Identification of therapeutic targets for quiescent, chemotherapy-resistant human leukemia stem cells. Sci Transl Med. 2010;2:17ra19

    CAS  Article  Google Scholar 

  52. 52.

    Taussig DC, Pearce DJ, Simpson C, Rohatiner AZ, Lister TA, Kelly G, et al. Hematopoietic stem cells express multiple myeloid markers: implications for the origin and targeted therapy of acute myeloid leukemia. Blood. 2005;106:4086–92.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  53. 53.

    Goardon N, Marchi E, Atzberger A, Quek L, Schuh A, Soneji S, et al. Coexistence of LMPP-like and GMP-like leukemia stem cells in acute myeloid leukemia. Cancer Cell. 2011;19:138–52.

    CAS  Article  PubMed  Google Scholar 

  54. 54.

    Hosen N, Park CY, Tatsumi N, Oji Y, Sugiyama H, Gramatzki M, et al. CD96 is a leukemic stem cell-specific marker in human acute myeloid leukemia. Proc Natl Acad Sci USA. 2007;104:11008–13.

    CAS  Article  PubMed  Google Scholar 

  55. 55.

    Tavakkoli M, Devlin SM, Park CY. CD99 is a therapeutic target on disease stem cells in acute myeloid leukemia and the myelodysplastic syndromes. Blood. 2013;122:2891.

    Google Scholar 

  56. 56.

    Barreyro L, Will B, Bartholdy B, Zhou L, Todorova TI, Stanley RF, et al. Overexpression of IL-1 receptor accessory protein in stem and progenitor cells and outcome correlation in AML and MDS. Blood. 2012;120:1290–8.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  57. 57.

    Kikushige Y, Shima T, Takayanagi S, Urata S, Miyamoto T, Iwasaki H, et al. TIM-3 is a promising target to selectively kill acute myeloid leukemia stem cells. Cell Stem Cell. 2010;7:708–17.

    CAS  Article  PubMed  Google Scholar 

  58. 58.

    Taussig DC, Miraki-Moud F, Anjos-Afonso F, Pearce DJ, Allen K, Ridler C, et al. Anti-CD38 antibody–mediated clearance of human repopulating cells masks the heterogeneity of leukemia-initiating cells. Blood. 2008;112:568–75.

    CAS  Article  Google Scholar 

  59. 59.

    Sarry JE, Murphy K, Perry R, Sanchez PV, Secreto A, Keefer C, et al. Human acute myelogenous leukemia stem cells are rare and heterogeneous when assayed in NOD/SCID/IL2Rgammac-deficient mice. J Clin Invest. 2011;121:384–95.

    CAS  Article  PubMed  Google Scholar 

  60. 60.

    Eppert K, Takenaka K, Lechman ER, Waldron L, Nilsson B, van Galen P, et al. Stem cell gene expression programs influence clinical outcome in human leukemia. Nat Med. 2011;17:1086–93.

    CAS  Article  Google Scholar 

  61. 61.

    Körbling M, Anderlini P. Peripheral blood stem cell versus bone marrow allotransplantation: does the source of hematopoietic stem cells matter? Blood. 2001;98:2900–8.

    Article  PubMed  Google Scholar 

  62. 62.

    Admiraal R, van Kesteren C, Jol-van der Zijde CM, Lankester AC, Bierings MB, Egberts TC, et al. Association between anti-thymocyte globulin exposure and CD4 + immune reconstitution in paediatric haemopoietic cell transplantation: a multicentre, retrospective pharmacodynamic cohort analysis. Lancet Haematol. 2015;2:e194–203.

    Article  PubMed  Google Scholar 

Download references


We thank the patients for participating in research that could not benefit them but only future patients. We thank the healthy volunteers involved in the study. This study could not happen without the dedication of Mamta Kantharia, Jennifer LeBlanc, Lori Rackel, Laura Spilchen, many inpatients nurses, pharmacists, particularly Michelle Dowhan, and physicians, notably Dr. Michelle Geddes, Dr. Mona Shafey, Dr. Peter Duggan, and Dr. Lynne Savoie. We also thank the staff of Calgary Laboratory Services, including Glenis Doiron. Finally, we thank Douglas Mahoney for invaluable feedback during this study.

Author contributions

R.D. developed the cytotoxicity assays, ran the assays, performed statistical analysis, and analyzed the results. P.D.K. and J.L. provided input into assay development and interpretation. M.M., T.v.S., and J.B. collected or arranged for the collection of specimens. A.D., D.M., F.M.K., and J.S. provided critical feedback. R.D. and J.S. designed the study, and J.S. supervised its conduct. R.D. wrote the manuscript.

Author information



Corresponding author

Correspondence to Rosy Dabas.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Electronic supplementary material

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Dabas, R., Dharmani-Khan, P., Modi, M. et al. Anti-thymocyte globulin’s activity against acute myeloid leukemia stem cells. Bone Marrow Transplant 54, 549–559 (2019).

Download citation

Further reading


Quick links