The Effect of Chemotherapy on Stroke Risk in Cancer Patients
Cancer is associated with an increased risk of thromboembolic events (TEEs).[1 ]
[2 ]
[3 ]
[4 ]
[5 ]
[6 ] The underlying mechanisms of cancer-associated thrombosis are complex because many
factors can contribute to TEE, including site and stage of the cancer, type of treatment,
such as chemotherapy, and patient characteristics, such as age ([Fig. 1 ]).[7 ] Cancer type-specific mechanisms include tissue factor-positive extracellular vesicles
for pancreatic cancer and podoplanin expression for brain cancer.[8 ]
[9 ]
[10 ] Therefore, it is difficult to determine the relative contribution of these different
factors to thrombosis. In this issue of Thrombosis and Haemostasis , Kitano et al analyzed the effect of chemotherapy on stroke in cancer patients.[11 ]
Fig. 1 Risk factors for cancer-associated thrombosis. Risk factors for thrombosis in cancer
patients include tumor characteristics, treatment, and patient characteristics. Tissue
factor (TF)-positive extracellular vesicles (EV), podoplanin, leukocytosis, and thrombocytosis
may enhance thrombosis in cancer patients.
The rate of venous thromboembolism (VTE) (1–19%) in cancer is much higher than the
rate of arterial thromboembolism (ATE) (0–5%).[1 ]
[2 ]
[3 ]
[4 ]
[5 ] One study found that the 6-month cumulative incidence of ATE (composite of myocardial
infarction and ischemic stroke) and ischemic stroke was significantly increased in
cancer patients compared with controls patients (ATE 4.7% vs. 2.2%; ischemic stroke
3.0% vs. 1.6%).[12 ] The risk of ATE in cancer patients was affected by cancer stage and to a lesser
extent by cancer type (with the highest rate for lung cancer). Another study also
found an increase in the 3-month cumulative incidence of ischemic stroke was higher
in patients with cancer compared to controls and affected by cancer type (lung 5.1%,
pancreatic 3.4%, colorectal 3.3%, breast 1.5%, and prostate 1.2%).[13 ] Cerebral infarction was also observed in nonsmall cell lung cancer patients (2.9%)
with those with brain metastasis having the highest rate (6.3%).[14 ] Gastric cancer patients are also prone to ischemic stroke after surgery.[15 ] Stroke patients with cancer have a worse prognosis compared with stroke patients
without cancer.[14 ]
[16 ]
[17 ]
Numerous chemotherapeutic agents are used to treat various forms of cancer that include
untargeted “conventional” agents, such as cisplatin, and targeted “unconventional”
agents, such as tyrosine kinase inhibitors. Cisplatin-based chemotherapy was shown
to be associated with a high rate of TEE (18.1%) in 932 patients with a variety of
cancers, but most of these events were VTEs with only 1.5% of the events being ATEs.[18 ] Another study with bladder cancer patients found that patients treated with platinum-based
chemotherapy had a significantly higher rate of TEE compared with patients who did
not receive chemotherapy (19.5% vs. 11.6%).[19 ] Several studies have investigated the effect of both conventional and nonconventional
chemotherapy on stroke in cancer patients ([Table 1 ]). One study investigated the effect of chemotherapy and/or radiotherapy on stroke
in head and neck cancer patients and concluded that patients < 55 years of age but
not patients ≥55 years of age had an increased risk of stroke with chemotherapy, radiotherapy,
or both compared with patients with surgery alone.[20 ] Another study concluded that chemotherapy, especially platinum-based regimes, was
an independent risk factor for stroke in ovarian cancer patients.[21 ] In contrast, chemotherapy (cisplatin or carboplatin) did not increase the risk of
stroke in patients with stage II to III bladder cancer.[19 ] Other studies have investigated the effect of nonconventional chemotherapeutic agents
on stroke. For instance, the vascular endothelial growth factor inhibitor bevacizumab
increased the overall relative risk of cerebrovascular events in patients with a variety
of cancers by 3.28.[22 ] In addition, use of the vascular endothelial growth factor receptor tyrosine kinase
inhibitors sunitinib and sorafenib was associated with an increase in stroke in patients
with renal cell carcinoma.[23 ]
Table 1
Studies investigating the association between chemotherapy and stroke in cancer patients
Study
Cancer type
Chemotherapy type
Total no. of patients
No. of patients received chemotherapy (%)
No. of stroke in patients without chemotherapy (%)
No. of stroke in patients received chemotherapy (%)
Association between chemotherapy and stroke
Ref
Kuan et al
Ovarian
Various types
8,810
6,590 (74.8)
N/A
N/A
Yes
[21 ]
Huang et al
Head and neck
Platinum
10,172
663 (under the age of 55) (6.5)
42 (surgery only) (2.5)
24 (3.6)
Yes
[20 ]
Gupta et al
Bladder
Platinum
5,057
1,079 (21.3)
216 (5.4)
54 (5.0)
No
[19 ]
Zuo et al
Various types
Bevacizumab
12,705
6,421 (50.5)
14 (0.2)
59 (0.9)
Yes
[22 ]
Jang et al
Kidney
Tyrosine kinase inhibitors
1,458
670 (46.0)
N/A
24 (3.6)
Yes
[23 ]
Kitano et al
Various types
Various types
19,007
5,887 (31)
51 (0.39)
44 (0.75)
Yes/No[a ]
[11 ]
a The association was not significant after adjusting for cancer status.
Kitano et al[11 ] analyzed the effect of chemotherapy on stroke by comparing the rates in patients
with (n = 5,887) or without (n = 13,119) chemotherapy. The study included a variety of cancer types and both conventional
(9–60% of patients) and nonconventional (6–14% of patients) chemotherapeutic agents.
Cancer patients who received chemotherapy had a higher rate of stroke (0.75%) compared
with patients who did not receive chemotherapy (0.39%). Kaplan–Meier curve analysis
of the data indicated a significant difference between the two groups (hazard ratio
1.84; 95% confidence interval 1.23–2.75). Importantly, however, there was no significant
difference after adjustment for cancer status (cancer stage and site). In addition,
chemotherapy was also not associated with increased stroke after adjustment for cancer
status in either a stratified Cox regression model or a time-dependent covariate Cox
regression model. Similarly, subanalysis indicated platinum-based chemotherapy did
not increase stroke.
The strengths of the study by Kitano et al[11 ] are its size and the adjustment for cancer status and age. However, there are some
limitations. For instance, the study includes a variety of cancer types, which are
known to have a different incidence of stroke, and various types of conventional and
nonconventional chemotherapies, which have a different impact on the risk of stroke.
The authors did not perform subanalysis of chemotherapy other than platinum-based
regimes and did not perform subanalysis of cancer type because the number of events
in each cancer type was too small.
In conclusion, Kitano et al concluded that chemotherapy is not associated with increased
risk of stroke in a general cancer patient population after adjustment for cancer
status. However, future studies are needed to investigate the effect of specific classes
of agents in specific cancer types.
