Thromboembolic events in brain tumour patients on bevacizumab

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ionc20
Acta Oncologica
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/ionc20
Thromboembolic events in brain tumour patients
on bevacizumab
Gunjesh Kumar Singh , Nandini Menon , Monica Madhusing Jadhav , Rutuja
Walavalkar , Hollis DSouza , Somnath Roy , Sudeep Das , Sujay Srinivas , Dilip
Harindran Vallathol & Vijay M. Patil
To cite this article: Gunjesh Kumar Singh , Nandini Menon , Monica Madhusing Jadhav , Rutuja
Walavalkar , Hollis DSouza , Somnath Roy , Sudeep Das , Sujay Srinivas , Dilip Harindran Vallathol
& Vijay M. Patil (2020): Thromboembolic events in brain tumour patients on bevacizumab, Acta
Oncologica
To link to this article: https://doi.org/10.1080/0284186X.2020.1815834
Published online: 08 Sep 2020.
Submit your article to this journal
View related articles
View Crossmark data

LETTER TO THE EDITOR
Thromboembolic events in brain tumour patients on bevacizumab
Gunjesh Kumar Singh, Nandini Menon, Monica Madhusing Jadhav, Rutuja Walavalkar, Hollis DSouza,
Somnath Roy, Sudeep Das, Sujay Srinivas, Dilip Harindran Vallathol and Vijay M. Patil
Department of Medical Oncology, Tata Memorial Hospital, Mumbai, India
ARTICLE HISTORY Received 26 May 2020; Accepted 21 August 2020
Background
Venous thromboembolism (VTE) is a frequently encountered
phenomenon and is also seen in 1–2% of normal population
[1]. However, it is a more common entity in post-operative
period where impaired mobility leads to significant venous
stasis. Apart from this, hypercoagulable state and injury to
the endothelial lining are the two other proposed mecha-
nisms responsible for its development [2,3]. Cancer patients
are especially vulnerable to VTE and an increase in the level
of tissue factor (pro-coagulant) is believed to be the possible
explanation to this [4]. However, brain tumours deserve spe-
cial mention due to the greater incidence of VTE in them.
According to Stein et al.VTE was seen in 3.5% of all hospital-
isations for brain tumour [5]. Similarly in another study by
Smith et al.the incidence in this group was 19.4% [6]. Many
authors have explored and proposed various factors that
have link with its development. Sartori et al. suggested that
the circulating tissue factor does not have a contributory
role in causing VTE in all cancers, but in high-grade brain
tumours, it certainly is one of the important precipitating fac-
tors [4]. Petterson et al.found age and sex to be the associ-
ated risk factors for VTE in this population [7]. Besides these,
obesity, paresis, residual tumour, chemotherapy and anti vas-
cular endothelial growth factor (VEGF) treatment have also
been listed as significant predisposing factors [8]. Heenkenda
et al. evaluated the role of genetic and non-genetic factors
in causation of VTE among glioblastoma multiforme (GBM)
patients and found B blood group to be a predictive risk fac-
tor [9].
Bevacizumab alone or in combination with chemotherapy
has been approved by United States Food and Drug
Administration (FDA) in recurrent and progressive GBM see-
ing the improved outcomes [10]. However, the European
Medicines Agency (EMA) does not recommend its use and
hence bevacizumab is the standard of care in these patients
in the United States but not in Europe [11]. VTE is one of the
most well known and dreaded side effects of this drug [12].
Khorana et al.worked in this direction, developed and vali-
dated a score to assess the risk of VTE, taking primary
tumour site; pre-chemotherapy haemoglobin level, platelet
counts, total leukocyte count (TLC) and body mass index
(BMI) as components [13]. However, they couldn’t draw any
conclusion for brain tumour as a risk factor for VTE because
of insufficient number of patients (n¼4).
To our belief, the use of bevacizumab in brain tumour
patients does add to the risk of VTE in them. Hence, we con-
ducted this study aiming to calculate the risk of VTE in this
group of patients and its association with Khorana score.
Materials and method
Selection of patients
The primary brain tumour patients undergoing systemic ther-
apy since 1
st
July 2015 in neuro-medical oncology unit of
Tata Memorial Hospital, Mumbai, India, were included in this
retrospective analysis. The eligibility criteria were:
1. Adult ambulatory patient aged 18 years
2. Relapsed or progressive GBM –WHO 2016 classification
of brain tumours was used for the diagnosis. WHO 2007
classification was used for patients diagnosed
before 2016.
3. Treated with bevacizumab
4. Time period –1
st
July 2015 to 31
th
December 2018
The patients fulfilling the above mentioned inclusion crite-
ria completely were selected and the data was entered in an
excel sheet. The study methodology was approved by the
Institutional Ethics Committee –III, Advanced Centre for
Treatment, Research and Education in Cancer (ACTREC),
Mumbai 410210, India. Waiver of consent was obtained.
Principle of declaration of Good clinical practice (GCP) and
International Council for Harmonisation of Technical
Requirements for Pharmaceuticals for Human Use (ICH) were
obtained. All patients had left written informed consent prior
to chemotherapy.
Data collection
From the above mentioned data set, VTE events, the respect-
ive Khorana scores along with the demographic data were
extracted and following were noted –
CONTACT Vijay M. Patil vijaypgi@gmail.com Department of Medical Oncology, Tata Memorial Hospital, Parel 400012, Mumbai, India
Co-first authorship.
ß2020 Acta Oncologica Foundation
ACTA ONCOLOGICA
https://doi.org/10.1080/0284186X.2020.1815834

Occurrence of VTE –Clinical suspicion (symptomatic) was
confirmed by Doppler imaging or computed tomography
angiogram. VTE occurrence during and 30 days after the
last bevacizumab dose was taken into account.
Concurrent use of chemotherapy and steroids
Pre-bevacizumab haemoglobin level.
Pre-bevacizumab platelets count
Pre-bevacizumab TLC level
BMI of the patient
Khorana score was calculated and patients were divided
into 3 risk groups [13]. As per this scoring system, scores
were allotted to pre-bevacizumab laboratory and clinical
parameters: primary tumour site (score 1 or 2), haemoglobin
leveL <100 g/L or red cell growth factor use (score-1); TLC
>11 10
9
/L (score-1); platelet count 350 10
9
/L (score-1);
BMI 35 kg/m2 (score-1). As in our study, the primary
tumour site was brain and Khorana score does not include
this tumour as high risk, we used score 0 for the first cat-
egory. All the scores were added to get a final score based
on which the final risk stratification was done as:
1. Low risk –score 0
2. Intermediate risk –score 1 or 2
3. High risk –score 3
Treatment
All the patients with relapsed or progressive GBM were dis-
cussed in joint neuro-oncology meeting and asked for inclu-
sion in the study. These patients were ineligible for re-
surgery or re-irradiation and hence were offered bevacizu-
mab either alone or in combination with chemotherapy. The
choice of single agent therapy or combination was based on
Eastern Cooperative Oncology Group Performance Status
and the presence of co-morbidities. Bevacizumab was admin-
istered every 2–3 week. Each dose was given as per standard
protocol. Premedication drugs like ondansetron (8 mg), dexa-
methasone (12 mg) and ranitidine (50 mg) were used
15 20 min prior to bevacizumab. Bevacizumab in a dose of
10 mg/kg body weight was infused over 90 min in the first
cycle and if tolerated well subsequent doses were adminis-
tered in 60 min. The drug was continued either till disease
progression or intolerable side effects. The drug was also
stopped if financials of the patient were inadequate.
Statistical analysis
Statistical analysis was done via SPSS version 20 and R ver-
sion 3.5.3. Descriptive analysis was performed. Pearson cor-
relation analysis was done and the Pearson correlation
coefficient was estimated between concurrent chemotherapy
use, steroid use, Khorana score and risk of VTE. p-Value .05
was considered significant.
Results
Baseline characteristics
Out of total 80 patients, 60 (75%) were males and 20 (25%)
were females. 68 (85%) patients belonged to the younger
and 12 (15%) belonged to the older (age >60 years) group.
The median number of bevacizumab cycles given was 5
(range, 1–24) and median duration of follow up was
26 months (range, 1–37 months).
VTE during bevacizumab and assosciated factors
Out of the 80 patients included in the study; 7 (8.8%) had
VTE events after starting bevacizumab. VTE was diagnosed in
the form of deep vein thrombosis (DVT) in 4 (5%) and pul-
monary thromboembolism (PTE) in 3 (3.8%) patients. Out of
these 7 patients, 2 expired and the cause of death was
believed to be PTE. 3 (42.8%) patients received concurrent
chemotherapy with irinotecan while 1 (14.28%) received
lomustine. All patients with VTE were given steroids (dexa-
methasone) during the course of treatment. We didn’t find
any significant association between VTE and the concurrent
use of chemotherapy (Fisher exact test, p-value ¼.387);
meanwhile use of steroid was also not seen to influence the
risk of VTE, (Fisher exact test, p-value ¼.685).
Khorana score and its correlation with VTE
Of all, 43 (53.8%) patients belonged to the low risk category
of Khorana score, 37 (46.2%) to the intermediate category
and none to the high risk category. Table 1 mentions
detailed Khorana scores. Out of 7 patients with VTE, 2
(28.57%) patients had Khorana score 0, 4 (57.14%) patients
had score 1 and only 1(14.28%) patient had score 2 (Table
2). There was no significant association between Khorana
scores obtained and VTE (fisher exact test, p-value ¼.171).
Discussion
This retrospective study was conducted to find out the inci-
dence of VTE in brain tumour patients undergoing anti VEGF
therapy (bevacizumab). It was seen in 8.8% of the enrolled
cases. We didn’t find any significant association between VTE
and the use of concurrent chemotherapy (Fisher exact test,
p-value ¼.387) and use of steroid was also not seen to
increase the risk of VTE (Fisher exact test, p-value ¼.685).
Further, we also noted that most of our patients diagnosed
with VTE belonged to either low risk or intermediate risk cat-
egory and none belonged to the high risk one.
Yust-Katz et al. conducted a retrospective study of 440
GBM patients to see the prevalence of VTE and its associ-
ation with the Khorana score. In contrast to our finding, they
found a higher occurrence of VTE in there cohort, which was
22%. They also proposed that obesity, recurrent VTE, raised
TLC and steroid use were the factors significantly associated
with the development of VTE in GBM [14]. Also, they labelled
Khorana score as a non-predictive tool for VTE diagnosis [14].
Similarly, in our study we also didn’t find any relation
2 G. K. SINGH ET AL.

between khorana score and VTE. However, our results may
be hypothesis-generating rather than definitive, due to the
small number of patients included. Misch et al.carried out a
retrospective study to see the rate of VTE in glioma patients
receiving bevacizumab and found VTE in 13% of patients.
They suggested that raised D-dimer levels and paresis were
the factors impacting VTE [15]. Kuk et al. did a retrospective
analysis to see VTE events in ovarian cancer patients receiv-
ing bevacizumab with or without chemotherapy and found
that VTE was evident only in the patients on bevacizumab.
They divided their patients into low, medium and high risk
categories as per Khorana score. Out of 57 ovarian cancer
patients, only 3 and 2 patients with high and medium risk
respectively experienced VTE. None of the patients with low
risk had VTE. They concluded that the use of bevacizumab
was associated with a statistically non-significant increase in
VTE risk in the high-risk group compared to the medium risk
group [16].
Many studies have also discussed the role of VTE prophy-
laxis in brain tumour patients knowing the obvious risk. In
this context, the role of anticoagulation therapy for preven-
tion of VTE in patients with high grade glioma was explored
in PRODIGE trial [17]. This therapy was seen to reduce VTE
events however, it was not statistically significant. Also, due
to the higher incidence of intracranial haemorrhage, the role
of prophylactic anticoagulation therapy remains uncertain in
these patients [17].
Patients with primary brain tumour have considerable risk
of developing VTE and the use of bevacizumab further adds
to the threat. However, the overall occurrence of VTE was
lower in our cohort in comparison to other studies in the lit-
erature. Hence we suggest development and instillation of a
VTE assessment tool for cancer patients in general and brain
tumour patients in particular, which can aid in easy and
timely recognition of the high risk cases so that prophylactic
anticoagulant therapy can be initiated.
There are some limitations of our study. First is the retro-
spective nature of the study. Our hospital is a tertiary centre
and majority of patients stay far and many come from other
states, which may cause under reporting of VTE as many
patients default and never report back. Hence a prospective
study will take care of this issue and help in better under-
standing. Secondly, the total number of cases included in
the analysis is small. Also, as only symptomatic or suspected
VTE patients are selected for further evaluation and confirm-
ation, the estimated rate of VTE gets compromised and may
not be representative. Still, to the best of our knowledge,
this is the first study from India attempting to find out the
incidence of VTE in brain tumour patients on bevacizumab
and hence is unique.
Conclusion
The incidence of VTE in primary brain tumour patients on
bevacizumab therapy is low. Concurrent use of chemother-
apy and steroids does not have an impact on the occurrence
of VTE. Low and intermediate risk Khorana scores are unable
to predict the risk of VTE in our population.
Ethical approval
The study methodology was approved by Institutional Ethics Committee
–III, Advanced Centre for Treatment, Research and Education in Cancer
(ACTREC), Mumbai 410210, India. Waiver of consent was obtained.
Principle of declaration of Good clinical practice (GCP) and International
Council for Harmonisation of Technical Requirements for
Pharmaceuticals for Human Use (ICH) was obtained. All patients gave
informed written consent before participation.
Disclosure statement
The authors declare that they have no competing interests.
Data availability statement
The authors confirm that the data supporting the findings of this study
are available within the article.
Table 1. Khorana score.
Parameters Number Percentage (%)
Pre-bevacizumab haemoglobin <10 g/dL or using RBC growth factors
Yes 5 6.3
No 75 93.7
Pre-bevacizumab platelet count 350 10
9
/L
Yes 3 3.8
No 77 96.3
Pre- bevacizumab leukocyte count >11 10
9
/L
Yes 09 11.3
No 71 88.7
Body mass index (BMI) 35 kg/m
2
Yes 25 31.3
No 55 68.7
Khorana score
0 43 53.8
1 33 41.3
245
300
Table 2. Khorana score and VTE.
Khorana score
Venous thromboembolism (VTE)
Yes No
0 2 (4.7%) 41 (95.3%)
1 4 (12.1%) 29 (87.9%)
2 1 (25%) 3 (75%)
300
ACTA ONCOLOGICA 3

References
[1] Fowkes FJ, Price JF, Fowkes FG. Incidence of diagnosed deep
vein thrombosis in the general population: systematic review. Eur
J Vasc Endovasc Surg. 2003;25(1):1–5.
[2] Wakefield TW, Myers DD, Henke PK. Mechanisms of venous
thrombosis and resolution. Arterioscler Thromb Vasc Biol. 2008;
28(3):387–391.
[3] Saha P, Humphries J, Modarai B, et al. Leukocytes and the natural
history of deep vein thrombosis: current concepts and future
directions. Arterioscler Thromb Vasc Biol. 2011;31(3):506–512.
[4] Sartori MT, Della Puppa A, Ballin A, et al. Prothrombotic state in
glioblastoma multiforme: an evaluation of the procoagulant activity
of circulating microparticles. J Neurooncol. 2011;104(1):225–231.
[5] Stein PD, Beemath A, Meyers FA, et al. Incidence of venous
thromboembolism in patients hospitalized with cancer. Am J
Med. 2006;119(1):60–68.
[6] Smith TR, Nanney AD, 3rd, Lall RR, et al. Development of venous
thromboembolism (VTE) in patients undergoing surgery for brain
tumors: results from a single center over a 10 year period. J Clin
Neurosci. 2015;22(3):519–525.
[7] Petterson TM, Marks RS, Ashrani AA, et al. Risk of site-specific
cancer in incident venous thromboembolism: a population-based
study. Thromb Res. 2015;135(3):472–478.
[8] Perry JR. Thromboembolic disease in patients with high-grade gli-
oma. Neuro-oncology. 2012;14(Suppl 4):iv73–80.
[9] Heenkenda MK, Malmstr€
om A, Lysiak M, et al. Assessment of gen-
etic and non-genetic risk factors for venous thromboembolism in
glioblastoma –the predictive significance of B blood group.
Thromb Res. 2019;183:136–142.
[10] Friedman HS, Prados MD, Wen PY, et al. Bevacizumab alone and
in combination with irinotecan in recurrent glioblastoma. J Clin
Oncol. 2009;27(28):4733–4740.
[11] Gil-Gil MJ, Mesia C, Rey M, et al. Bevacizumab for the treatment
of glioblastoma. Clin Med Insights Oncol. 2013;7:123–135.
[12] Nalluri SR, Chu D, Keresztes R, et al. Risk of venous
thromboembolism with the angiogenesis inhibitor bevacizu-
mab in cancer patients: a meta-analysis. JAMA. 2008;300(19):
2277–2285.
[13] Khorana AA, Kuderer NM, Culakova E, et al. Development and
validation of a predictive model for chemotherapy-associated
thrombosis. Blood. 2008;111(10):4902–4907.
[14] Yust-Katz S, Mandel JJ, Wu J, et al. Venous thromboembolism
(VTE) and glioblastoma. J Neurooncol. 2015;124(1):87–94.
[15] Misch M, Czabanka M, Dengler J, et al. D-dimer elevation and
paresis predict thromboembolic events during bevacizumab ther-
apy for recurrent malignant glioma. Anticancer Res. 2013;33(5):
2093–2098.
[16] Kuk A, Magnowska M, Suchy W, et al. Retrospective evaluation of
thromboembolism risk in ovarian cancer patients treated with
bevacizumab. Target Oncol. 2017;12(4):495–503.
[17] Perry JR, Julian JA, Laperriere NJ, et al. PRODIGE: a random-
ized placebo-controlled trial of dalteparin low-molecular-
weight heparin thromboprophylaxis in patients with newly
diagnosed malignant glioma. J Thromb Haemost. 2010;8(9):
1959–1965.
4 G. K. SINGH ET AL.