The treatment of cancer-associated stroke is determined based on stroke etiology; however, the optimal treatment and therapeutic benefits remain unclear. We aimed to evaluate the clinical and functional outcomes of patients with cancer and acute ischemic stroke, especially focusing on patients with cryptogenic stroke. We retrospectively reviewed consecutive cancer patients diagnosed with acute ischemic stroke at our institution between January 2018 and December 2019. Stroke subtype, cancer treatment after stroke, modified Rankin Scale (mRS) scores before and at ischemic stroke, 3 months after ischemic stroke or last follow-up, and discharge destination were evaluated. We analyzed 48 cancer patients with acute ischemic stroke, including 24 with cryptogenic stroke and 24 with known stroke mechanisms. The median survival time of the patients was 62 days. Thirty-one patients (64.6%) discontinued cancer treatment after stroke. Forty patients (83.3%) and 37 (77.1%) were in poor functional states (defined as mRS score ≥ 3) at stroke onset and 3 months after stroke or last follow-up. Twenty-two patients (91.7%) with cryptogenic stroke discontinued cancer treatment after stroke, whereas 15 patients (62.5%) with known stroke mechanisms continued cancer treatment (p = 0.0004). Home or rehabilitation hospital discharge destinations were less frequently seen in patients with cryptogenic stroke (n = 7, 29.2%) than those with known stroke mechanisms (n = 15, 62.5%, p = 0.021). Stroke has a significant negative impact on patients’ functional states and cancer treatment strategy, leading to short survival times, especially in patients with cryptogenic stroke.
Article
Assessment of clinical and functional outcome in patients with cancer and acute ischemic stroke
https://doi.org/10.21203/rs.3.rs-2877830/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Stroke is a common comorbidity in cancer patients. An autopsy study reported that 14.6% of cancer patients had pathologic evidence of cerebrovascular disease, and 7.4% had clinical symptoms of stroke 1. Stroke is the second leading cause of death in patients with cancer after cancer. A prospective observational study showed that approximately 70% of cancer patients receiving chemotherapy died from the progression of underlying cancer, whereas 9.2% died from stroke, equal to that of infection 2. Another report showed that the risk of fatal stroke in cancer patients was more than twice that in the general population 3. In addition, stroke is a major cause of disability, which is a serious issue because neurological disability may affect daily life and influence cancer treatment strategies 4. With recent advances in cancer treatment, the survival rate of cancer patients continues to increase 4. Therefore, managing stroke in patients with cancer is important for improving their outcomes.
A hypercoagulable state is common in patients with cancer 5–7. The pathophysiology of cancer-associated hypercoagulability is complex and involves multiple factors. Biological factors that promote thromboembolic events in cancer patients include tumor cell-derived procoagulant factors, inflammatory cytokines, fibrinolysis inhibitors, extracellular vesicles, neutrophil extracellular trap formation, increased platelet aggregation, and excessive endothelial adhesiveness 6. Accumulating evidence has shown that cancer-associated hypercoagulability is a major contributor to ischemic stroke in patients with cancer, particularly cryptogenic stroke 5,8.
The treatment of ischemic stroke in patients with cancer is determined individually based on the etiology of the stroke, underlying cancer, risk of bleeding, patients’ overall prognosis, and expressed goals of care 4,9. Hypercoagulability has been the main focus in treating patients with cancer classified as a cryptogenic or embolic stroke of undetermined source (ESUS); however, the optimal treatment remains undefined. Anticoagulant therapy with low-molecular-weight heparin (LMWH) is recommended for treating venous and arterial thromboses in patients with cancer 9. Direct oral anticoagulants (DOAC) have been reported to be effective against cancer-associated venous thromboembolism (VTE) and are used in patients with cancer and arterial and venous thrombosis 6,9,10. However, it remains unclear whether this evidence of DOAC for VTE could be extrapolated to treating ischemic stroke because the underlying mechanisms might differ 11. In addition, few clinical studies have evaluated the efficacy of DOAC against ischemic stroke in patients with cancer 6,10,12,13. Although anticoagulation therapies can effectively correct hypercoagulability in patients with cancer-associated stroke, the potential benefits of anti-stroke treatments on clinical outcomes have not been well defined 14.
This study aimed to investigate the clinical and functional outcomes of cancer patients with acute ischemic stroke who received anti-stroke treatment at our institution based on the etiology of stroke, focusing on patients with cryptogenic stroke. We also evaluated the potential benefits of anticoagulation therapy using heparin and DOAC in patients with cryptogenic stroke.
Patient characteristics
We identified 48 patients with cancer (0.044%) and acute ischemic stroke among 108,410 patients who underwent cancer treatment at our institution during the study period. The rate of acute ischemic stroke per 100,000-person was 44.2.
Patient characteristics are summarized in Table 1. The population comprised 23 men (47.9%) and 25 women (52.1%), with a median age of 70. Lung and pancreatic cancers were the most frequent cancers (n = 9, 18.8%), followed by hepatobiliary cancer (n = 7, 14.6%) and colorectal cancer (n = 5, 10.4%). Most patients were classified as having active cancer (n = 39; 81.3%). Twelve patients received anticoagulation therapy before stroke because of deep vein thrombosis (n = 7), previous ischemic stroke (n = 3), portal vein stent insertion (n = 1), and atrial fibrillation (n = 1) (Table 1).
Among the 48 patients, 24 (50.0%) were classified as cryptogenic, eight as cardioembolism (16.7%), eight as small vessel occlusion (16.7%), six as atherosclerosis (12.5%), and two as other determined etiologies (4.2%) (Table 1). After ischemic stroke onset, 45 patients (93.8%) received stroke treatment according to the stroke subtype, including edaravone (n = 37, 77.1%), unfractionated heparin (n = 23, 47.9%), antiplatelets (n = 12, 5.0%), DOAC (n = 16, 33.3%), warfarin (n = 2, 4.2%), and embolectomy (n = 2, 4.2%). Of the 24 patients with cryptogenic stroke, 17 (70.8%) received intravenous unfractionated heparin, and 10 (41.7%) received DOAC following intravenous unfractionated heparin. All the patients received edoxaban as a DOAC (Table 2). One patient had chronic subdural hematoma during antiplatelet therapy. Age, sex, cancer status, and prior antithrombotic treatment did not differ between patients with cryptogenic stroke and those with known stroke mechanisms. Lung, pancreatic, hepatobiliary, colorectal, gastric, breast and gynecologic cancers were more frequently observed in patients with cryptogenic stroke (n = 23, 95.8%) than in those with known stroke mechanisms (n = 12, 50.0%) (p = 0.0007). The median plasma D-dimer level was significantly higher in patients with cryptogenic stroke than those with known stroke mechanisms (21.2 vs. 3.0 ㎍/mL, P < 0.0001). The number of patients who received chemotherapy at stroke onset was 17 (70.8%) in cryptogenic stroke and 10 (41.7%) in known stroke mechanisms (small vessel occlusion, n = 4; atherosclerosis, n = 2; cardioembolism, n = 2; other determined etiologies, n = 2) (p = 0.042) (Table 1).
Clinical and functional outcomes
Of the 48 patients, six (12.5%) experienced a recurrent stroke, 31 (64.6%) discontinued cancer treatment after stroke, and 34 (70.8%) died at last follow-up (Table 3). The median survival time (MST) after stroke onset was 62 days. While 33 patients (68.8%) had mRS scores < 3 before stroke onset, 40 (83.3%) and 37 (77.1%) had mRS scores ≥ 3 at stroke onset and at 3 months after stroke or last follow-up, respectively (Fig. 1). The discharge destinations included home (n = 17, 35.4%), rehabilitation hospital (n = 5, 10.4%), hospice care (n = 12, 25.0%), and death (n = 14, 29.2%) (Table 3).
The MST was significantly shorter in patients with cryptogenic stroke than in those with known stroke mechanisms (34 vs. 590 days, p = 0.0008) (Fig. 2A). Most patients with cryptogenic stroke discontinued cancer treatment after stroke (n = 22, 91.7%). In contrast, more than half of the patients with known stroke mechanisms continued treatment (n = 15, 62.5%, p = 0.0004). At stroke onset, the number of patients with mRS < 3 was not different between the cryptogenic and known stroke mechanisms groups (n = 3, 12.5% vs. n = 5, 20.8%, p = 0.44%), however, the number of patients with severe functional disability (defined as mRS ≥ 5) was higher in the group with cryptogenic stroke (n = 8, 33.3% vs. n = 2, 8.3%, p = 0.028) (Fig. 2B). At 3 months or last follow-up, patients with mRS ≥ 3 were also more in the group with cryptogenic stroke (n = 21, 87.5%) than in the group with known stroke mechanisms (n = 16, 66.7%), although there was no statistical significance (p = 0.086) (Fig. 2C). Improvement in the mRS score from stroke onset to 3 months after stroke or the last follow-up was observed in four patients (16.7%) with cryptogenic stroke and eight patients (33.3%) with known stroke mechanisms (p = 0.32) (Fig. 2D). Home or rehabilitation hospital discharge destinations were less frequently observed in the group with cryptogenic stroke (n = 7, 29.2%) than in the group with known stroke mechanisms (n = 15, 62.5%; p = 0.021) (Table 3).
We evaluated the clinical outcomes of 10 patients with cryptogenic stroke who received DOAC after intravenous unfractionated heparin administration. Seven patients started treatment with edaravone and intravenous unfractionated heparin, then switched to DOAC; two had intravenous unfractionated heparin, then switched to DOAC; and one initially received intravenous unfractionated heparin, and DOAC was added after recurrent stroke. Two patients (20.0%) had recurrent strokes: one during unfractionated heparin treatment and the other during DOAC treatment. The MST of the 10 patients was 61 days. Only one patient (10.0%) continued cancer treatment. Nine patients had decreased mRS scores at stroke onset compared to pre-stroke mRS scores, and two patients (20.0%) had improved mRS scores from stroke onset to 3 months after stroke or last follow-up (Fig. 2E).
This study showed that more than 80% of patients were in a poor functional state (defined as mRS score ≥ 3) after stroke. The survival and functional outcomes of anti-stroke treatment were poor. In particular, cryptogenic stroke was significantly associated with shorter survival time, more frequent discontinuation of cancer treatment, and less frequent home or rehabilitation hospital discharge destinations than stroke with known mechanisms. We also showed that anticoagulant therapy with unfractionated heparin and DOAC had limited clinical and functional benefits in patients with cryptogenic stroke.
The management of ischemic stroke in patients with cancer is usually determined based on the stroke subtype, underlying cancer status, overall prognosis, and expressed goals of care 4,9; however, the outcome is generally poor 15–17. The MST from ischemic stroke ranges from 84 days to 4.5 months 15–17. Lee et al. showed that the MST from ischemic stroke was 109 days, and the mortality rate was 18.3% at 1 month and 71.6% at 1 year in patients with active cancer who had an acute ischemic stroke 16. The stroke recurrence rate in patients with cancer and acute ischemic stroke is 13.6%, and the cumulative rates of recurrent ischemic stroke have been reported to be 7% at 1 month, 13% at 3 months, and 16% at 6 months, respectively 17. The median mRS score at hospital discharge was 3 17 and the 3-month mRS score ≥ 3 was 75.6% 18. Our study observed that MST from ischemic stroke was 62 days, recurrent ischemic stroke rate was 12.5%, and mRS score ≥ 3 at 3 months after stroke or last follow-up was 77.1%, which confirmed the poor outcome in cancer patients with ischemic stroke.
Approximately 40–51% of ischemic strokes in patients with cancer are classified as cryptogenic or ESUS 5,8,11,19,20. The outcome of cryptogenic stroke in patients with cancer is poor 11,20. MST from ischemic stroke was reported to be 55–62 days 11,19,20. The stroke recurrence rate in patients with cancer and cryptogenic stroke is 58% 10. In our study, cryptogenic stroke was associated with frequent lung/pancreatic/hepatobiliary/colorectal/gastric/breast/gynecologic cancers, high serum D-dimer levels, and frequent chemotherapy. The frequency of cryptogenic stroke in patients with cancer was 50%, and the MST from ischemic stroke was 34 days, similar to previous reports 11,20.
A key finding of this study was that 91.7% of patients with cryptogenic stroke discontinued cancer treatment, whereas 62.5% of patients with known stroke mechanisms continued. This striking difference might be partially explained by the fact that patients with cryptogenic stroke had a more severe functional decline (Fig. 2B) and were less likely to show improvement (Fig. 2D) than those with known stroke mechanisms. Naito et al. also showed that among 26 patients with cancer-associated stroke, malignancy treatment was switched to palliative treatment in 11 (42.3%) after ischemic stroke onset 12. These observations suggest that stroke has a critical impact on functional state and trigger the discontinuation of cancer treatment, especially in patients with cryptogenic stroke. Primary prevention is important for the continuation of cancer treatment.
We also found that only 29.2% of the patients with cryptogenic stroke got home or rehabilitation hospital discharge destinations. In contrast, 62.5% of those with known stroke mechanisms got home or rehabilitation hospital discharge destinations. This difference may reflect the short survival time and discontinuation of cancer treatment in patients with cryptogenic stroke. Navi et al. showed that 56% of patients with cryptogenic stroke were discharged home, which is better than our results 20. Their better functional state might explain this difference at discharge; the median mRS score at discharge was 3, whereas ours was 6. Our results also suggested that stroke affects the discharge destination more in patients with cryptogenic stroke than in those with known stroke mechanisms.
Most cryptogenic stroke in patients with cancer is associated with cancer-associated hypercoagulability 6,7. Hasegawa et al. classified 87.3% of cryptogenic strokes as cancer-associated 19. Cancer-associated hypercoagulability is considered to be a spectrum of diseases ranging from thrombosis induced primarily by the production of tissue factors by tumor cells to a platelet-rich microthrombotic process triggered by carcinoma mucins and involving P- and L-selectins 7, suggesting that cancer-associated stroke involves the activation of both the coagulation system and platelet function.
Our treatment strategy for cryptogenic stroke aimed to initially correct hypercoagulability using unfractionated heparin and then prevent recurrent stroke using DOAC. Two of the 10 patients who underwent this treatment had recurrent stroke: one occurred during unfractionated heparin and the other during DOAC treatment. Nam et al. reported that cardio-cerebrovascular disease recurrence rates were 49% with LMWH treatment and 57% with DOAC treatment in patients with cryptogenic stroke and active cancer 10. Compared to their results, our results might be better; however, careful interpretation is needed because of the small sample size and short survival time. Functional recovery was observed in only two patients 3 months after stroke or at the last follow-up (Fig. 2E), suggesting the limited clinical effect of our treatment strategy. As DOACs are convenient to use, they would be the preferred choice of the anticoagulant with reduced treatment burden. Further studies are required to define the role of DOAC in treating cancer-associated stroke.
Our study has certain limitations. First, it was a retrospective study with small sample size. Therefore, our results must be confirmed in larger studies. Second, we did not perform precordial or transesophageal echocardiography in any patient. Thus, the assessment of the cardioembolic sources may have been underestimated in this study. Third, we performed MRI and confirmed stroke recurrence only when neurological symptoms were observed. Therefore, the rate of recurrent stroke may have been underestimated in this study.
We showed poor survival and functional outcomes in patients with cancer and acute ischemic stroke. Stroke has a significant negative impact on patients’ functional states. It triggers the discontinuation of cancer treatment, leading to short survival times and less frequent discharge to home or rehabilitation hospitals, especially in patients with cryptogenic stroke. Anticoagulant therapy with unfractionated heparin and DOAC has limited clinical and functional benefits in patients with cryptogenic stroke. Further studies are needed to be carried out to present preventive measures against ischemic stroke in patients with cancer and to develop an optimal treatment for cancer-associated stroke.
Patients
This was a retrospective, observational study. We analyzed consecutive patients with acute ischemic stroke treated at the National Cancer Center Hospital between January 2018 and December 2019. The inclusion criteria were as follows: (1) documented acute ischemic stroke based on diffusion-weighted imaging (DWI) and (2) patients with any cancer who were treated at our institution before stroke onset. Patients with primary central nervous system (CNS) tumors were included in this study. Patients with hemorrhagic stroke were excluded from the study.
Clinical assessment
The clinical and radiological records of the patients were reviewed. Data on the following variables were collected: age, sex, type of cancer, date of cancer diagnosis, clinical history and cancer treatment before ischemic stroke, clinical history, and cancer treatment after ischemic stroke, modified Rankin Scale (mRS) scores (before and at ischemic stroke, and 3 months after ischemic stroke or last follow-up), discharge destination, plasma D-dimer level at the onset of ischemic stroke, date of death or last hospital visit, bleeding complications, and the date of ischemic stroke recurrence and death. Recurrent ischemic stroke was a new lesion detected by magnetic resonance imaging in patients with new acute neurological symptoms. Active cancer was defined as a cancer diagnosis or treatment for cancer during the 6 months preceding stroke diagnosis or the presence of recurrent or metastatic cancer 8,13.
Ischemic stroke assessment and management
The ischemic stroke subtypes were classified according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria as large-artery atherosclerosis, cardioembolism, small-vessel occlusion, stroke of other determined etiology, and stroke of undetermined etiology 21. In this study, stroke of undetermined etiology was termed cryptogenic stroke 22. The treatment for ischemic stroke was determined based on the stroke mechanism. The neuroprotective drug edaravone was used to minimize ischemic brain damage, irrespective of stroke etiology, except in patients with poor functional status or severe renal dysfunction 23. Patients with large-artery atherosclerosis and small-vessel occlusion were treated with antiplatelet agents, those with cryptogenic stroke were treated with initial intravenous unfractionated heparin for 1–2 weeks, followed by DOAC, and those with cardioembolism were treated with DOAC. The treatment decision was made individually based on the patient’s medical condition, physical performance, underlying cancer status, and overall patient prognosis. We did not use LMWH as an anticoagulant because it has not been approved for stroke treatment in Japan.
Routine evaluations included magnetic resonance imaging (MRI), magnetic resonance (MR) angiography, electrocardiography (ECG), and 24-h ECG monitoring. Echocardiography and carotid ultrasonography was not performed routinely. Blood tests, including those for D-dimer levels, were also carried out.
Statistical analysis
Age and serum D-dimer levels between patients with cryptogenic stroke and those with known stroke mechanisms were evaluated using the Mann–Whitney U test. Sex, cancer type (lung/pancreatic/hepatobiliary/colorectal/gastric/breast/gynecologic vs. glioblastoma/primary CNS lymphoma/others), cancer status (active vs. inactive), prior antithrombotic treatment (yes vs. no), cancer treatment at stroke onset (chemotherapy vs. other therapy), mRS score at and before stroke (< 3 vs. ≥ 3), mRS score 3 months after stroke or last follow-up (< 3 vs. ≥ 3), cancer treatment after stroke (continue vs. discontinue), and discharge destination (home or rehabilitation hospital vs. hospice or death) were evaluated using the χ2 test or the Fisher’s exact test. The survival time from stroke onset was defined as the interval between the date of stroke onset and death or censorship at the last follow-up. These were calculated using the Kaplan–Meier method and compared using the log-rank test. Analyses were conducted using JMP® 16.2.0 (SAS Institute Inc., Cary, NC, USA) and GraphPad Prism® version 9.5.0 (GraphPad Software, La Jolla, CA, USA). In all tests, statistical significance was set at p < 0.05.
Funding: The authors declare that no funds, grants, or other support was received during the preparation of this manuscript.
Competing interests: The authors have no relevant financial or nonfinancial interests to disclose.
Author Contributions: Y.N. designed the study. M.O., Y.M., M.T., S.Y., Y.O., H.Y., T.O., N.S., T.M., Y.K., and Y.N. contributed to the data collection, analysis, and interpretation. M.O. conducted the statistical analyses. M.O. and Y.N. wrote the first draft of the manuscript, which was edited by all authors. All the authors have read and approved the final version of the manuscript.
Data Availability: The datasets used and/or analyzed in the current study are available from the corresponding author on reasonable request.
Ethics approval: All procedures performed in this study were in accordance with the ethical standards of the Institutional Review Board and the 1964 Declaration of Helsinki and its later amendments. This study was approved by the internal review board of the National Cancer Center (approval number: 2020-178).
Consent to participate: Written informed consent was obtained from all individual participants.
Consent to publish: No individual personal data is contained in this manuscript.
- Graus, F., Rogers, L. R. & Posner, J. B. Cerebrovascular complications in patients with cancer. Medicine (Baltimore). 64, 16-35 (1985).
- Khorana, A. A., Francis, C. W., Culakova, E., Kuderer, N. M. & Lyman, G. H. Thromboembolism is a leading cause of death in cancer patients receiving outpatient chemotherapy. J Thromb Haemost. 5, 632-634 (2007).
- Zaorsky, N. G. et al. Stroke among cancer patients. Nat Commun. 10, 5172 (2019).
- Lun, R., Siegal, D., Ramsay, T. & Dowlatshahi, D. Cancer and stroke: What do we know and where do we go? Thromb Res. 219, 133-140 (2022).
- Kim, S. G. et al. Ischemic stroke in cancer patients with and without conventional mechanisms: a multicenter study in Korea. Stroke. 41, 798-801 (2010).
- Navi, B. B. et al. Cancer and Embolic Stroke of Undetermined Source. Stroke. 52, 1121-1130 (2021).
- Varki, A. Trousseau's syndrome: multiple definitions and multiple mechanisms. Blood. 110, 1723-1729 (2007).
- Seok, J. M. et al. Coagulopathy and embolic signal in cancer patients with ischemic stroke. Ann Neurol. 68, 213-219 (2010).
- Gladstone, D. J. et al. Canadian Stroke Best Practice Recommendations: Secondary Prevention of Stroke Update 2020 - ADDENDUM. Can J Neurol Sci, 1 (2022).
- Nam, K. W. et al. Treatment of Cryptogenic Stroke with Active Cancer with a New Oral Anticoagulant. J Stroke Cerebrovasc Dis. 26, 2976-2980 (2017).
- Shin, Y. W. et al. Predictors of survival for patients with cancer after cryptogenic stroke. J Neurooncol. 128, 277-284 (2016).
- Naito, H. et al. Antithrombotic Therapy Strategy for Cancer-Associated Ischemic Stroke: A Case Series of 26 Patients. J Stroke Cerebrovasc Dis. 27, e206-e211 (2018).
- Nakajima, S. et al. Post-Treatment Plasma D-Dimer Levels Are Associated With Short-Term Outcomes in Patients With Cancer-Associated Stroke. Front Neurol. 13, 868137 (2022).
- Kleindorfer, D. O. et al. 2021 Guideline for the Prevention of Stroke in Patients With Stroke and Transient Ischemic Attack: A Guideline From the American Heart Association/American Stroke Association. Stroke. 52, e364-e467 (2021).
- Cestari, D. M., Weine, D. M., Panageas, K. S., Segal, A. Z. & DeAngelis, L. M. Stroke in patients with cancer: incidence and etiology. Neurology. 62, 2025-2030 (2004).
- Lee, M. J. et al. Hypercoagulability and Mortality of Patients with Stroke and Active Cancer: The OASIS-CANCER Study. J Stroke. 19, 77-87 (2017).
- Navi, B. B. et al. Recurrent thromboembolic events after ischemic stroke in patients with cancer. Neurology. 83, 26-33 (2014).
- Cutting, S., Wettengel, M., Conners, J. J., Ouyang, B. & Busl, K. Three-Month Outcomes Are Poor in Stroke Patients with Cancer Despite Acute Stroke Treatment. J Stroke Cerebrovasc Dis. 26, 809-815 (2017).
- Hasegawa, Y., Setoguchi, T., Sakaida, T. & Iuchi, T. Utility of a scoring system for differentiating cancer-associated stroke from cryptogenic stroke in patients with cancer. Neurol Sci. 41, 1245-1250 (2020).
- Navi, B. B. et al. Cryptogenic subtype predicts reduced survival among cancer patients with ischemic stroke. Stroke. 45, 2292-2297 (2014).
- Adams, H. P., Jr. et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke. 24, 35-41 (1993).
- Hart, R. G. et al. Embolic strokes of undetermined source: the case for a new clinical construct. Lancet Neurol. 13, 429-438 (2014).
- Fidalgo, M. et al. Edaravone for acute ischemic stroke - Systematic review with meta-analysis. Clin Neurol Neurosurg. 219, 107299 (2022).
Table 1: Patient characteristics
Table 2: Summary of treatments
Table 3: Summary of clinical outcomes
No competing interests reported.