ORIGINAL ARTICLE

Eltrombopag therapy in newly diagnosed steroid non-responsiveITP patients

Anil Kumar Tripathi • Ayush Shukla •

Sanjay Mishra • Yogendra Singh Yadav •

Deependra Kumar Yadav

Received: 30 October 2013 / Revised: 22 January 2014 / Accepted: 29 January 2014

� The Japanese Society of Hematology 2014

Abstract Primary immune thrombocytopenia (ITP) is an

autoimmune disorder characterised by isolated thrombo-

cytopenia (peripheral blood platelet count \100 9 109/L)

in the absence of other causes or disorders that may be

associated with thrombocytopenia. The upfront treatment

in newly diagnosed ITP patients is steroids; however, about

one-third patients do not respond, and require other treat-

ment, including IVIg, anti-D, or splenectomy. Previous

studies have shown decreased platelet production in some

ITP patients, aside from the evidence of enhanced platelet

destruction. Thrombopoietin receptor agonists (TPO-RA),

such as eltrombopag have been shown to provide good

response in steroid non-responsive chronic ITP patients.

We have studied response to eltrombopag in 25 newly

diagnosed steroid non-responsive ITP patients; 80 %

patients showed response at the end of 1 month, and 76 %

sustained response at the end of 3 months. The platelet

count rose from a mean value of 17.5 ± 3.6–152.5 ±

107.9 9 109/L at the end of 1 month. Our results suggest a

possible role of eltrombopag in newly diagnosed steroid

non-responsive ITP patients. However, our study is limited

in that it is a single-centre study, with a small sample size,

and lacks a long-term safety profile. Our findings highlight

the potential value of a larger prospective study on the

upfront use of TPO-RA in patients of ITP.

Keywords Primary ITP � ITP � TPO-RA � Eltrombopag

Introduction

Primary immune thrombocytopenia (ITP) is an autoim-

mune disorder characterised by isolated thrombocytopenia

(peripheral blood platelet count \100 9 109/L) in the

absence of other causes or disorders that may be associated

with thrombocytopenia [1]. The diagnosis of ITP remains

one of exclusion; no robust clinical or laboratory parame-

ters are currently available to establish its diagnosis with

accuracy. The main clinical problem of ITP is an increased

risk of bleeding, although bleeding symptoms may not

always be present. ITP is said to be ‘‘newly diagnosed’’ if

time of presentation from the diagnosis is within 3 months

and chronic if lasting for more than 12 months [1].

The disease is characterised by enhanced destruction of

antibody-coated platelets in the reticuloendothelial system.

In some patients, there is also an evidence of immune-

mediated megakaryocytic suppression/injury [2]. The main

aim of the treatment is to avoid bleeding by augmenting the

low platelet counts that may have caused bleeding.

As there is no consensus for etiology, similar is the

situation with treatment. Treatment strategies have focus-

sed on limiting the accelerated platelet destruction. Steroid

is the initial treatment choice but 1/3 patients do not

respond or need intolerable, larger doses [3]. The mecha-

nism by which steroid acts is not very clear, however, it is

supposed to be through its immunosuppressant effect on T

cells. In patients not responding to steroids, various treat-

ment options are splenectomy, anti-D, IV immunoglobulin,

Rituximab, danazol, vincristine, etc. [3].

Thrombopoietin receptor agonists (TPO-RA: oral el-

trombopag, subcutaneous romiplostim) have been tried in

patients with ITP on the basis that there is megakaryocytic

suppression/injury in some, if not all, patients of ITP.

Previous studies have shown that eltrombopag increases

A. K. Tripathi � A. Shukla (&) � S. Mishra �Y. S. Yadav � D. K. Yadav

Department of Clinical Hematology and Medical Oncology,

King George’s Medical University, Lucknow 226003,

U.P., India

e-mail: ayushgeorgian@gmail.com

123

Int J Hematol

DOI 10.1007/s12185-014-1533-y

the platelet counts significantly in patients of chronic ITP

and reduces the morbidity and mortality associated with the

disease [4]. These studies have focussed on the role of

TPO-RA in patients of chronic ITP who failed steroid

therapy or other first-line treatment. However, there is a

paucity of published research work on the role of TPO-RA

in newly diagnosed (time of diagnosis within 3 months)

ITP patients who have failed steroid therapy. We hereby

prospectively studied the response to eltrombopag in newly

diagnosed ITP patients non-responsive to steroid and

sought to know the durability of response on TPO-RA.

Methods

This prospective cohort study comprised 27 patients of

primary ITP from the hematology clinic who fulfilled the

inclusion criteria such as:

1. ‘‘Newly diagnosed ITP’’ (time from diagnosis within

3 months).

2. Non-responsive to steroid at 2 weeks.

3. Could not opt for IVIg, anti-D or splenectomy as

further treatment.

The study was approved by the institutional ethical

committee and all patients had given a written informed

consent before enrolment. Non-responsiveness to steroid

was defined according to international working group

consensus [1].

Cases of ITP associated with secondary causes like HIV,

SLE, H. pylori, other cancer and pregnancy were excluded

from the study.

Cases underwent detailed clinical history, physical

examination and investigations. A complete hemogram

with thorough peripheral blood smear examination and

tests to exclude secondary causes of ITP including viral

profile (HIV, HBsAg, HCV), H. pylori, anti-thyroid, and

ANA and APLA were done.

Patients of newly diagnosed ITP who did not respond to

steroid therapy were given eltrombopag. Initial dose was

50 mg/day for 1 month which was tapered to 25 mg/day if

platelet count was sustained [150 9 109/L for a week.

Steroid was tapered in 2 weeks after eltrombopag was

started. Platelet count was done biweekly for first month,

and then monthly for 3 months. Decision to stop therapy

was taken if counts rose above 250 9 109/L in order to

reduce the thrombocytosis-associated risk. If there was no

response at 1 month or anytime later, patients were offered

choice of other modalities like splenectomy, anti-CD 20,

etc.

The primary end point was defined as a response to

eltrombopag that is platelet count more than 50 9 109/L at

the end of 1 and 3 months of treatment. The secondary end

points included safety and tolerability, affordability and

signs of bleeding. The incidence and severity of bleeding

were assessed at every visit according to WHO bleeding

scale.

The patients were followed fortnightly. The clinical

course including symptoms and side effects was recorded

on a regular follow-up.

Results

Out of newly diagnosed ITP patients (time from diagnosis

within 3 months) attending OPD/IPD clinics, only those

(n = 27) patients were included in the present study who

did not respond to steroid. Two patients were lost during

follow-up and there results are not included. The mean age

of patients (n = 25) was 27 ± 8.63 years. All patients

were 15 years old or above (range 15–55 years). Males

constituted 40 % (10/25) of the study population. The

mean platelet count at presentation was 13.7 ± 5.2 9 109/

L, median 12 9 109/L and the range was 20 9 109/L.

Clinical features included petechiae and purpura in 68 %,

ecchymosis in 40 % and mucosal bleeding in 60 % of

patients. All patients were newly diagnosed (time from

diagnosis within 3 months) cases of primary ITP and were

treatment-naive. All patients were treated in outpatient

clinic or day care centre. Indications for treatment were

bleeding in 76 %, platelet count B10 9 109/L in 8 %, and

both features in 16 %.

Patients who were included in the study had already

received prednisolone at a dose of 2 mg/kg/day for

2 weeks did not show response nor chose to go for sple-

nectomy or other therapy such as IVIg or anti-D. The mean

platelet count in these patients at the end of 2 weeks of

steroid treatment was 17.5 ± 3.6 9 109/L, median

18 9 109/L and the range was 16 9 109/L. These patients

were given 50 mg/day of eltrombopag, and steroid was

gradually tapered and discontinued within 2 weeks.

At 2 weeks of eltrombopag, the mean value of platelet

counts was 152 9 109/L (range 10–470 9 109/L). At

1 month of eltrombopag therapy, the primary end point, as

the response to eltrombopag, was seen in 80 % (20 out of

25) patients. Response was 70 % in males and 86.8 % in

females. Mean platelet count at 1 month of eltrombopag

therapy was 152.5 ± 107.9 9 109/L, median 150 9 109/L

and range (11–450 9 109/L) (Fig. 1). The durable response

at 3 months was seen in 76 % of the patients with a mean

platelet count of 141.8 ± 44.9 9 109/L, median

120 9 109/L and range (10–225 9 109/L) (Fig. 2). Twelve

of 25 patients had attained platelet counts [150 9 109/L

and their eltrombopag was reduced to 25 mg/day. We did

not stop eltrombopag even in those with good response

within 3 months, as previous studies have shown risk of

rebound thrombocytopenia on stopping it.

A. K. Tripathi et al.

123

Five patients who did not respond to eltrombopag 50 mg

and one patient who lost response subsequently were

offered other options along with splenectomy.

No significant difference in response to eltrombopag

between male and female population was noted. Two

patients had platelet count of more than 200 9 10-9 but

did not withdraw from the study.

Bleeding episodes and severity decreased significantly

in all those who achieved response (76 %). Bleeding

symptoms were assessed by the WHO scale.

Six of our patients (24 %) had mild headache on el-

trombopag treatment but did not necessitate any treatment.

No patient was reported with episode of thrombosis. Three

of our 25 patients (12 %) had mild elevation in ALT

(\60 IU) and hence did not require any intervention. The

values returned to normal within 1 month.

Discussion

ITP is characterised by isolated thrombocytopenia (periph-

eral blood platelet count \100 9 109/L) in the absence of

other causes or disorders that may be associated with

thrombocytopenia [1]. The upfront choice of therapy is ste-

roid. Various studies in past have reported response to steroid

in 53–80 % patients [3]. Patients who do not respond to

steroid are given additional therapy like IVIg or anti-D and

those who are non-responsive are candidates for

splenectomy. The manner in which steroid acts is not ade-

quately known; however, it is thought to be by suppression of

T cell and antibody-mediated destruction of platelets [3].

Why one-third patients of ITP do not respond to steroid is not

precisely known. It may be assumed that these patients may

have other prominent pathogenic mechanisms for thrombo-

cytopenia in addition to increased platelet destruction [5–9].

Houwerzijl et al. [2] have reported evidence of megak-

aryocytic suppression or injury contributing to thrombo-

cytopenia in patients of ITP. This has formed the basis for

the use of newer agents that accelerate production of

Fig. 1 Comparison of mean

platelet counts during treatment

0102030405060708090

100

At one month At 3 month

80 76

20 24

%

Treatment duration

Responders Non-responders

Fig. 2 Comparison of response at the end of 1 and 3 months of

eltrombopag treatment

Eltrombopag in newly diagnosed ITP

123

platelets. TPO-RA such as eltrombopag and romiplostim

have been tried in patients with chronic ITP and phase II/III

trials have shown a good response with adequate safety and

efficacy [10–12]. Conventionally, in the steroid non-

responsive newly diagnosed patients, the choice is IVIg or

splenectomy. In our setup, most patients are either not able

to afford IVIg (average cost of therapy being $2500) or are

not willing to undergo splenectomy due to culturally

inherent fear of surgery. Also splenectomy has its own

limitations. It is effective only in 60–75 % of patients and

is associated with an increase in the risk to thrombosis.

Hence splenectomy in newly diagnosed ITP may be a hasty

decision. Moreover, few patients of adult ITP may attain

spontaneous remission once the effects of severe or

symptomatic thrombocytopenia are taken care of and

splenectomy can be avoided in them [3]. Eltrombopag has

previously been used effectively in chronic ITP patients

who were non-responsive to steroid or other agents [4, 10,

11]. However, its role in newly diagnosed ITP patients is

yet undefined. There is need for newer agents to be used in

newly diagnosed ITP patients who do not respond to ste-

roids. These agents should be able to increase the platelet

count within days in order to cover the period of bleeding

or severe thrombocytopenia. This justifies the use of TPO-

RA in newly diagnosed ITP patients. According to current

practice guideline, TPO-RA is not initial treatment of

choice in either naive, newly diagnosed patients or patients

non-responsive to steroid. Safety of TPO-RA in its short

term and to some extent long term uses is now established

by various studies. In addition, the response to eltrombopag

is visible within few days and the response is durable.

Hence, we sought to explore the role of eltrombopag in

newly diagnosed ITP patients (diagnosed within 3 months)

who were non-responsive to steroid. We used eltrombopag

as it has convenient oral dosing and is easily available.

We gave eltrombopag in 25 newly diagnosed patients of

ITP who did not show response after 2 weeks of steroid

therapy. The response to eltrombopag was observed in

80 % patients at 1 month. There was no effect of initial

platelet counts on the response to eltrombopag (Table 1).

There are no published studies on the effect of eltrombopag

in newly diagnosed ITP patients to the best of our

knowledge; however, when compared with a large study in

chronic ITP, where response was seen in 81 % patients,

response in our study was found to be comparable [4].

We tapered the dose of eltrombopag to 25 mg daily in

the responders and offered other options as splenectomy to

the non-responders. At the end of 3 months, response was

seen to be persisting in 76 % patients, which was higher

than the response seen in patients of chronic ITP as studied

by Cheng and Saleh [10, 11]. So the response to el-

trombopag in newly diagnosed patients can be considered

to be more durable than in patients of chronic ITP.

The mechanism by which TPO-RA increases platelet

count in patients not responding to steroid is a matter of

further studies. However, it is likely to be due to stimula-

tion of platelet production in such patients. In experimental

studies, TPO-RA was shown to stimulate production of

platelets in both the normal as well as injured bone marrow

[3, 13].

The pathogenetic mechanisms in ITP are heterogeneous

and the proposed mechanisms are (1) antibody-mediated

destruction, (2) T cell-mediated destruction, and (3)

megakaryocyte injury/suppression [5]. It is likely that one

or many mechanisms may operate in a single patient and the

response to therapy may depend upon the dominant

underlying mechanisms. We have previously shown that

patients who had evidence of bone marrow megakaryocytic

abnormality were less likely to respond to eltrombopag

[14]. Hence it can be assumed that the response to a par-

ticular agent may depend upon the predominant mechanism

underlying the pathogenesis of thrombocytopenia in ITP.

At present, the initial choice of therapy in newly diag-

nosed ITP patients is not individualised and steroid is

arbitrarily given to all patients of ITP who need treatment.

With addition of knowledge about different underlying

mechanisms of thrombocytopenia in ITP, it is imperative to

use other agents alone or in combination with steroids in

order to target different pathogenetic mechanisms. While

the specific ideal initial treatment is a matter of future

endeavours, combination therapy including steroid with

eltrombopag may be chosen as initial treatment in newly

diagnosed ITP patients, or eltrombopag may promptly be

chosen as agent of choice in case patients do not respond to

steroid within 2 weeks. Study population is biased because

it did not include other non-responders who opted for other

choices.

Our study has demerits in that it is a single-centre study,

has small sample size and lacks long-term safety profile.

This manuscript is intended to present our findings that

Table 1 Effect of initial

platelet count on response to

treatment with eltrombopag

Initial platelet count No. of patients At 1 month treatment At 3 months treatment

B10000 6 189000.00 ± 144391.10 133666.70 ± 74395.34

[10000 19 141052.60 ± 95715.71 105736.80 ± 63978.16

p value 0.35 0.37

A. K. Tripathi et al.

123

there is a possible role of eltrombopag in a subgroup of

newly diagnosed ITP patients and in no way it recommends

the use of eltrombopag in all newly diagnosed ITP patients.

However, the response to eltrombopag in newly diagnosed

steroid non-responsive ITP patients in the present study

sets a platform for a longer study on a larger number of

patients to substantiate the role of eltrombopag in these

patients.

Conflict of interest None.

References

1. Rodeghiero F, Roberto S, Terry GM, et al. Standardization of

terminology, definitions and outcome criteria in immune throm-

bocytopenic purpura of adults and children: report from an

international working group. Blood. 2009;113:2386–93.

2. Houwerzijl EJ, Blom NR, Van der Want JJL, et al. Ultrastructural

study shows morphologic features of apoptosis and para-apop-

tosis in megakaryocytes from patients with idiopathic thrombo-

cytopenic purpura. Blood. 2004;103:500–6.

3. Stasi R, Provan D. Management of immune thrombocytopenia in

adults. Mayo Clin Proc. 2004;79:504–22.

4. Bussel JB, Cheng G, Saleh MN, et al. Eltrombopag for the

treatment of chronic idiopathic thrombocytopenic purpura.

N Engl J Med. 2007;357:2237–47.

5. McMillan R. The pathogenesis of chronic immune thrombocy-

topenic purpura. Semin Hematol. 2007;44(4 Suppl 5):S3–11.

6. Olsson B, Anderson PO, Jemas M, et al. T cell mediated cyto-

toxicity toward platelets in chronic idiopathic thrombocytopenic

purpura. Nat Med. 2003;9(9):1123–4.

7. Chow L, Aslam R, Speck ER, et al. A murine model of severe

thrombocytopenia is induced by antibody and CD 8? T cell

mediated responses that are differentially sensitive to therapy.

Blood. 2010;115(6):1247–53.

8. Sakakura M, Wada H, Tawara I, et al. Reduced CD4? CD25- T

cells in patients with idiopathic thrombocytopenic purpura.

Thromb Res. 2007;120(2):187–93.

9. Ballem PJ, Segal GM, Stratton JR, et al. Mechanism of throm-

bocytopenia in chronic autoimmune thrombocytopenic purpura.

Evidence of both impaired platelet production and increased

platelet clearance. J Clin Invest. 1987;80(1):33–40.

10. Cheng G, Saleh MN, Marcher C, et al. Eltrombopag for man-

agement of chronic immune thrombocytopenia (RAISE): a

6-month, randomized, phase 3 study. Lancet. 2011;377(9763):

393–402.

11. Saleh MN, Bussel JB, Cheng G, et al. EXTEND Study Group.

Safety and efficacy of eltrombopag for treatment of chronic

immune thrombocytopenia: results of the long-term, open-label

EXTEND study. Blood. 2013;121(3):537–45.

12. Kuter DJ, Rummel M, Boccia R, et al. Romiplostim or standard

of care in patients with immune thrombocytopenia. NEJM.

2010;363(20):1889–961.

13. Erickson-Miller CL, Delorme E, Tian SS, et al. Preclinical

activity of eltrombopag (SB-497115), an oral, nonpeptide

thrombopoietin receptor agonist. Stem Cells. 2009;27(2):424–30.

14. Tripathi A, Mishra S, Yadav D et al. Megakaryocyte morphology

and its impact in predicting response to steroid in immune

thrombocytopenia. Platelets 2013 (in press).

Eltrombopag in newly diagnosed ITP

123