Haemopoietic stem cell transplants: the impact of haemorrhagic complications

Haemopoietic transplants: haemorrhagic complications Andrea

stem cell the impact of

count usually approaches normal values towards day 106 post-transplantation, many patients face a period of approximately 3 months during which they experience a clinically significant thrombocytopenia. The resulting increase in susceptibility to haemorrhage during this time represents a significant complication, which is often exacerbated by tissue damage and subsequent multi-organ problems, leading to a more pronounced bleeding tendency. As any haemorrhagic complications may thus prove to be fatal, successful management of bleeding following stem cell transplants is essential in order to achieve and maintain a favourable outcome.

Bacigalopo

Department

of Haematology,

Bone Marrow

Transplant

Unit, Ospedale. San

:

Martino, Geneva, Italy

Abstract Acute graft-versus-host disease (GVHD) is the most important complication of allogeneic haemopoietic stem cell transplantation (HSCT), increasing susceptibility to haemorrhage and risk of early mortality. We evaluated 807 allogeneic HSCT patients to assess both the association between bleeding and GVHD, and the influence of haemorrhagic complications on clinical outcome. Up to 55% of patients with grade Ill-IV GVHD experienced bleeding, compared with 23% of patients with grades O-l. Furthermore, 45% of patients receiving non-HLA-identical transplants suffered haemorrhage, whereas only 23% of patients receiving transplants from HLA-identical donors experienced bleeding. This can be explained by the higher incidence of severe GVHD among recipients of non-H& identical transplants. Our findings also demonstrated that haemorrhagic complications - particularly bleeding from the GI tract - markedly increase patient mortality. An ongoing, multi-centre, randomised, double-blind trial is currently investigating the efficacy and safety of recombinant factor Vlla (rFVlla; NovoSeven@) in the treatment of HSCTassociated bleeding. The trial will examine the ability of three dose levels of rFVlla to reduce - or even eliminate entirely the incidence and severity of haemorrhage following such procedures. A total enrolment of 100 patients is anticipated, and preliminary data are expected at the end of 2003. 0 2003 Elsevier Science Ltd. All rights reserved. KEY WORDS: haemorrhage;

haemopoietic graft-versus-host

stem cell transplant; disease; recombinant

allogeneic; factor Vlla

INTRODUCTION

aemopoietic stem cell transplants are complex therapeutic procedures in which healthy haemopoietic stem cells are given to individuals who have undergone chemotherapy for one of many forms of leukaemia, lymphoma, aplastic anaemia, and other disorders. A conditioning regimen of high-dose chemotherapy and/or radiotherapy is typically given over 5-10 days prior to the stem cell transplant, in order to eradicate diseased or cancerous cells. This conditioning regimen may also result in the destruction of normal, healthy cells in the bone marrow, as illustratedtby a dramatic reduction in platelet count (Fig. 1). Follow&g the stem cell transplant, recovery of platelet count is a slow and gradual process, with 50% of patients remaining thrombocytopenic at day 50.’ Although the platelet

Blood Reviews

(2003) f7, S&S IO

0 ZOO3 Ekvier

Science Ltd. All ri@s ~~.w&.

AUTOLOGOUS STEM CELL

AND ALLOGENEIC TRANSPLANTS

HAEMOPOIETIC

The number of autologous and allogeneic haemopoietic stem cell transplants performed in Europe has increased considerably since the 1970s. Just 16 allogeneic bone marrow transplants were undertaken in 1973, compared with 1353 10 years later. The next IO-year period witnessed not only a further increase in the number of transplants undertaken, but also a gradual change in donor type. Of the 7737 stem cell transplants undertaken in Europe in 1993,40% involved an allogeneic donor, and 60% were autologous.2 Just five years later in 1998, the number of stem cell transplants had risen to 20;892 (29% allogeneic; 71% autologous).3 Today, over 20,000 allogeneic and autologous stem cell transplants are performed each year in Europe. These data highlight the continuing increase in utilisation of stem cell transplants over the last 30 years, and reflect their role as a beneficial and effective therapeutic modality. Over the last three decades there has been an increase in the number of autologous stem cell transplantations performed, accompanied in more recent years by a change in the type of human leukocyte antigen (HLA) donor used in allogeneic transplantation. Although the number of allogeneic transplants using HIA-identical donors (i.e. siblings) has remained at around 3,000 per year in Europe since 1995, the number of transplants using non-HLA identical and alternative donors has risen. In 1995, 273 transplants using related, non-HIA-identical donors were performed, compared with 383 in 1998. The use of alternative, unrelated donors has shown a larger increase, with 752 and 1,407 stem cell trans plantations undertaken in 1995 and 1998, respectively.2 This increase in the use of unrelated donors may impact on the clinical outcome of allogeneic stem cell transplantation, as the potential of HLA-mismatched donors to increase the risk of postoperative immune and haemorrhagic complications could result in a more prominent predisposition to bleed.

GRAIT-VERSUS-HOST

DISEASE

The most important complication resulting from allogeneic stem cell transplantation is acute graft-versus-host dise&e (GVHD), which is the most significant factor influencing clinical outcome.4 Acute GVHD occurs when the donor’s T-lymphocytes recognise the recipient’s cells as foreign and mount an immune response to reject them. Subsequent

200 T 160--

HSCT

177

160--

-20?

0

20

50

Days from Fig. I. Reduction

in platelet

count

associated

with

100

160

365

HSCT

HSCT.

HSCT, haemopoietic

stem

cell transplant.

: erbated by platelet consumption resulting from the bleeding episodes commonly experienced by such patients. The re: suiting thrombocytopenia may last for several months. Graft : versus host disease can therefore cause clinically signiticant : morbidity and is the primary cause of death in patients : receiving allogeneic stem cell tmnsplants.8

; ; Severity Fig. 2. Severity plants.

GVHD.

of acute

GVHD

graft-versus-host

of GVHD in HIA-identical

(grade) and unrelated

trans-

disease.

activation of both prointlammatory cytokine cascades and immune effector cells5 results in a distinctive syndrome of dermatitis, hepatitis, and gastroenteritis, damaging target tissues and leading to an increased risk o! morbidity and early patient mortalhy6 Clinical classification of acute GVHD involves four stages tanging from grade I, which is characterised by milder symptoms, to grade III, with severe multi-organ involvement, and grade Iv, which may be fatal.’ Grades III and IV are more common in patients receiving stem cell transplants from unrelated donors, as the greater the HLA mismatch between donor and recipient, the higher the risk and intensity of GVHD (Pig. 2). This explains the higher incidence of immune and haemorrhagic problems in recipients of unrelated stem cell transplants. The severity of acute GVHD correlates directly with the platelet count. Patients with no evidence of GVHD - regardless of donor type - often regain normal platelet levels (approximately 150 x log/L) within 50 days of the stem cell transplant. In contmst, the platelet counts of patients with grade IV GVHD almost never fully recover. This may be attributable to the effect of inflammatory cytokines on haemopoietic function, and is likely to be exac-

JUEMORRHAGIC HAEMOPOIETIC

COMPLICATIONS IN ALLOGENEIC STEM CELL TRANSPLANTS

We evaluated 807 patients undergoing allogeneic haemopoietic stem cell transplants in order to determine the influence of haemorrhagic complications on clinical outcome, and to assess the association between bleeding and GVHD. The primary end-points of the study were the incidence, site, and clinical impact of haemorrhages. Conducted over a 10;year period between 1991 and 2001, the study involved an adult patient population (median age 35 years; range 16-66 years) receiving stem cells from an HLA-identical sibling (570 patients), a non-HLA-identical family member (74 patients), or an unrelated donor (163 patients). Bone marrow was used as a stem cell source in 701 patients, and peripheral blood was used as the source in the remaining 106 participants. Peripheral blood stem cell transplants are being increasingly used in the allogeneic context,’ and their inclusion in this study allowed a direct comparison of different stem cell sources. An extended follow-up period (median duration 576 days; range lo-4277 days) facilitated analysis and characterisation of any complications arising as a result of the transplant. Following stem cell transplantation, 259 (32%) patients suffered at least one bleeding episode with a mean duration of 7 days (range 5-9 days). Haemorrhagic cystitis was the most common manifestation of bleeding, followed by gastrointestinal (GI) haemorrhage, and bleeding from the mouth or nose. The majority of haemorrhagic patients experienced just one bleeding site, with very few patients bleeding from three or four different locations (Table 1).

0 2003 Elsevier Science ltd. All rights reserved.

Blood Reviews

(2003) 17, S&S IO

I Table I

Haemorrhagic

complications

following

allogeneic

HSCT

p = 0.001

p = 0.8

p=O.B

250 Haemorrhages

following

Number

HSCT

of patients ,i0

Number

of bleeding

200 1

piiiEij

sites 567

None One

196

Two

30

Three

I2

Four

2 ”

Site of haemorrhage Bladder

21

(haemorrhagic

cystitis)

135

GI tract

62

Mouth/nose

40

Uterus

I4

HSCT:

(menometrorrhagia)

haemopoietic

stem

cell transplant;

Fig. 4. Platelet blood

GI: gastrointestinal.

A further finding of this study demonstrated that, as expected, the incidence of haemorrhagic complications varied according to donor type. In total, 45% of patients receiving transplants from non-HLA-identical donors suffered from haemorrhage, compared with just 23% of patients receiving stem cells from an HLA-identical sibling (p = 0.0001). This can be explained by the higher incidence of a more severe GVHD among patients receiving non-HLA-identical transplants, resulting in thrombocytopenia and a subsequent increase in bleeding tendency. Our results indicate that bleeding is a highly signiticant complication of allogeneic stem cell transplants, with approximately one-third of patients experiencing haemorrhage. The inIluence of donor type on the frequency of bleeding complications is also highly signilicant and clinically relevant, with non-HLA-identical transplants resulting in a far greater incidence of haemorrhagic problems. This tinding is particularly relevant for today’s allogeneic stem cell transplants: as families in Europe become smaller, the need for unrelated donors is steadily increasing. Impact of GVHD and stem haemorrhagic complications Our study also assessed the incidence of haemorrhagic Fig. 3, patients with acute

cell source

on

impact of acute GVHD on the complications. As illustrated by GVHD demonstrated a higher

1 n No hemorrhage

q Hemorrhage

p = 0.03

no GVHD Fig. 3. Association GVHD,

Blood

graft-versus-host

Reviews

(2003)

of acute

GVHD GVHD

with

haemorrhagic

complications.

disease.

17, S&S/O

30 Time

0 2003 Eketier Science ltd. All rights reserved.

1

as a stem

count

post-transplant

following

HSCT

50 (days) using bone

marrow

or peripheral

cell source.

haemorrhagic tendency than those without GVHD. Therefore, the presence of GVHD is a significant predictor of bleeding complications, although the risk is also present in the absence of any immune complications. The severity of GVHD also correlates with the incidence of bleeding episodes; up to 55% of patients with grade III-IV GVHD suffered from haemorrhagic complications, compared with 23% and 31% of patients with grades O-I and II, respectively. These results are in accordance with several reports in the literature claiming that haemostatic disturbances are strongly associated with the presence of GVHD, which represents a significant risk factor for haemorrhagic complications following stem cell transplantation.‘O~l’ Examination of the infhtence of stem cell source on haemorrhagic complications revealed a slightly higher incidence of bleeding episodes in patients receiving bone marrow transplants when compared to those receiving stem cells from peripheral blood (24% and 18% of patients, respectively). However, this difference did not approach statistical significance (p = 0.3). This lack of a clear correlation between stem cell source and bleeding tendency may be explained by the rate of engraftment following stem cell transplantation. As demonstrated by the platelet count, engraftment at day 21 post-transplant is significantly more rapid with peripheral blood stem cells than with stem cells from bone marrow. However, from day 30 onwards, platelet counts are almost exactly the same in transplants from both sources (Fig. 4). This observation is overlooked in most reports in the literature, ‘* but it may explain why the current study failed to find a significant effect of stem cell source on the incidence of bleeding complications. Impact of haemorrhage on mortality Haemorrhagic complications impact significantly on survival rate following stem cell transplants. Kaplan-Meier analysis (Fig. 5) indicates that lo-year survival in patients suffering from haemorrhagic complications is just 50%, compared with 75% in patients who do not bleed. Preliminary analysis of haemorrhagic cystitis, the most common bleeding problem in the participating patiems, demonstrates that, by itself, this particular complication does not appear to impact significantly on survival rate. In contrast, GI bleeding is a significant predictor of reduced survival, resulting in a mortality rate of almost 80%. Patients

.

x.

_

i--^-.-----I---.---L

Hemorrhage2

p = 0.01

Fig. 5. Impact

of haemorrhage

who do not exhibit GI haemorrhage, however, demonstrate a much greater chance of survival in the years following stem cell transplantation, with a far lower mortality rate of 25%. These data reinforce the concept that effective management o{ haemorrhage in stem cell transplantation patients is essential in order to achieve a successful outcome and to reduce the incidence of early patient mortality. In addition, successful control of bleeding may influence important economic factors by reducing the duration of hospital stay. When compared to a median hospital stay of 52 days for patients who do not experience haemorrhagic complications, patients who bleed typically require a median hospitalisation of 70 days (p < 0.0001). At the current rate, one day in hospital costs C1,500, and so a hospital stay that is extended by approximately 12 days due to bleeding complications results in an additional cost of 618,000. A reduction in the incidence of haemorrhagic problems therefore has the potential to impact significantly on economic considerations. Study conclusions f In our study, one-third of patients undergoing allogeneic stem cell transplantation developed a bleed that lasted for approximately 7 days. In addition, the incidence of bleeding complications correlated with both the type of donor and the presence and severity of GVHD. This can be explained by the thrombocytopenia that commonly occurs in the presence of GVHD, which in turn is influenced by the type of HLA donor used. Interestingly, and in contrast with several reports in the literature, our study found no influence of stem cell source on the occurrence of bleeding problems. This finding may be due to the fact that both bone marrow stem cells and peripheral blood stem cells produced a very similar rate of haemopoietic recovery, despite a more rapid engraftment with peripheral blood until day 2 1. The haemorrhagic complications observed in this study impact significantly on patient mortality. As suggested in the literature, bleeding complications may not be the direct cause of death, but they have a potential prognostic impli-

on survival

rate.

cation as a predictor of a poor clinical outcome in the years following the transplant.‘3 Bleeding from the GI tract may be a particularly important predictor of transplant mortality, whereas bladder haemotrhages, manifested as haemorrhagic cystitis, do not significantly influence post-transplantation survival rate. In addition, the occurrence of bleeding following stem cell transplantation may significantly increase costs associated with the treatment, as patients experiencing haemorrhagic problems often require an extended hospitalisation period. Therefore, if haemorrhages could be effectively treated or reduced in duration, it may be possible to reduce the cost of allogeneic stem cell transplants by around 618,000 in approximately one-third of patients. As 6,000 allogeneic transplants are performed each year in Europe, an effective strategy to control bleeding would therefore result in highly significant economic benefits.

RECOMBINANT FACTOR VIIa (NovoSeven@) ALLOGENEIC STEM CELL TRANSPLANTS

IN

In an attempt to reduce the incidence of haemorrhagic complications following allogeneic stem cell transplants, Novo Nordisk (Denmark) has initiated a multicentre, randomised, double-blind study investigating the efficacy and safety of recombinant factor VIIa (rFVIIa; NovoSevenm) in the treatment of such bleeding. The objective of this ongoing parallel group, placebocontrolled trial is to evaluate the efficacy of placebo and three doses of NovoSeven@in patients undergoing allogeneic stem cell transplants. In addition, the safety of all treatment regimens used will be assessed. The primary end-point of the study is the effect of NovoSeven@on haemorrhage, as defined by a change in bleeding score during a period of 38 hours between visit 0 and visit 2. This short trial will therefore examine the ability of treatment to alter the incidence and severity of haemorrhage, and will categorise bleeding as stopped, decreased, unchanged, or worsened.

0 2003 Ekevier Science Ltd. All rights reserved.

Blood Reviews

(2003) 17, S&S IO

A secondary end-point is to assess patients’ transfusion requirements of red cells (number of units/ml), platelets (number of units/ml), and fresh frozen plasma (number of units/ml) over a 4day observation period. Bleeding will be evaluated at 24, 48, 72, and 96 hours post-transplantation. The study wilI therefore investigate whether treatment with NovoSeven@can effectively reduce the incidence of haemorrhage following stem cell transplantation, or even eliminate it entirely, as well as whether transfusion requirements of stem cell transplant patients can be altered.

References

I.

A,

transplant: 2.

Raiola

AM,

van

recovery

graft

versus

and cell dose.

Gratwohl

A. Hermans

etic stem

cell transplants

for Blood

Lint

after host

infections

Group

MT, et al. Factors

allogeneic

disease,

donor

type,

Br J Haematol200 J. Indications in Europe

and Bone Marrow

influenc-

haemopoietic

stem

cells

cytomegalovirus

I ; I 12: 2 19-227.

and donor

source

of hemopoi-

1993: report

from

the European

Transplantation.

Clin Transplant

1995; 9: 355-363. 3.

Trial design, inclusion criteria, and enrolment Patients enrolled into the study will be tandomised to receive either placebo plus standard transfusion therapy, or a repeated dose regimen using 40 @g/kg, 80 pg/kg, or 160 Kg/kg NovoSeven@plus standard transfusion therapy over a 2day period. Infusion of NovoSeven’%ill be repeated every 6 hours until 7 doses have been administered, in order to detect evidence of a dose response. To be included in the study, participants must have undergone an allogeneic stem cell transplant, be 212 years of age, and have provided informed consent. Only patients with active bleeding continuing for 3 consecutive days will be enrolled. Patients with ecchymoses, line oozing, or skin bleeding will excluded from the study, regardless of the duration of the bleeding episode. Since the trial began in May 2000, approximately 70 patients have been recruited; a total enrolment of 100 patients is anticipated. It is also expected that preliminary data will be available by the end of 2003. These data wiIl provide important information regarding the efficacy and safety of NovoSeven@in the management of bleeding following allogeneic stem cell transplants, and the study results are therefore eagerly awaited by patients and clinicians alike.

Dominietto

ing haematologic

Gratwohl

A, Passweg

Hematopoietic 2000; 4.

Gratwohl

A. Brand

6.

myeloid

leukemia. JLM.

E, Tiberghien

Flowers

Oncol

Darmstadt velopment

Frassoni

E van of stem

S, Enger

Pihusch tients.

Vogelsang

acute

A,

stem

cell therdisease.

107.

GB,

Farmer

ER, Horn

indicators

of the de-

graft-versus-host-disease.

2002;

Lint

87(Suppl

J Invest

on survival.

Transplantation

R, Salat C, Schmidt

H, Fahmy

D. Peripheral

blood

hematopoietic

stem

2002;

bleeding with

or periph-

transplantation.

after

graft

74: 1303-l

vs. bone

host

bone disease

1999; 67: 68 I-689.

a retrospective

0, Kamel

allogeneic

versus

E. et al. Hemostatic

transplantation:

Transplantation

Mahmoud

marrow

allogeneic

8): 4-8.

C, Swan V, et al. Acute association

marrow

MT. Bone

cells for

transplantation:

and effect

12.

a cytokine

of graft-versus-host

1999; 13: 109 l-l

AD,

as a source

bone

versus

disease:

and histopathologic

Bacigalupo

marrow I I.

graft

198.

KM. Hematopoietic

Am

of progressive

chronic

2000; 6: 66-74.

and treatment

eral blood Nevo

of acute I): I95-

graft-versus-host

Donnenberg

for

1992; 99: 397-402.

Haematologica IO.

76(Suppl

E, Sullivan

laboratory,

disease

transplantations

effecters

2002;

Clin North

CL,

Clinical,

Dermatol 9.

J

100: 3877-3886.

Hematologie

MED. Kansu

Hematol TD.

A.

1998. Hematol

J, et al. Graft-versus-host

2002;

I? The

model?

apy. Pathophysiology 8.

J, Urbano-lspizua

in Europe

sibling

and cytokine

Int J Hematol

dysregulation 7.

Blood

Cellular

disease.

Robinet

R, Apperley

in HLA-identical

Ferrara host

H. Hermans

I : 333-350.

and outcome 5.

J. Baldomero

stem cell transplantation

of 447

in pa-

309.

A, Kamel marrow

complications analysis

M, El-Haddad as a source

cell transplantation.

Bone

S, Swan V, Enger C, et al. Acute

bleeding

A, El-Kadi

for

Marrow

allogeneic Transplant

1999; 24: 355-358. 13.

Nevo

transplantation tative

Blood Reviews

(2003)

17, S&S IO

0 2003 Efsevier Science Ltd. All rights reserved.

analysis

(BMT)

-

incidence

in 1,402 patients.

and effect

Blood

after

bone

on survival.

1998; 9 I : I469-

marrow A quanti-

1477.