Something else to add to your doctor's stroke protocols; checking for cancer.
YOUR DOCTOR'S RESPONSIBILITY. Or I suppose you can take on that and give your doctor a pass.
Ischemic stroke: A paradoxical manifestation of cancer
Keywords
1. Introduction
Stroke is a heterogeneous pathologic process that results in acute neurologic injury. Cancer is one of the many risk factors associated to it. Globally, both stroke and cancer, represent a significant public health burden. In the specific case of Colombia, both are leading causes of death, stroke in the second place and cancer occupying the third place (Stefan et al., 2009; Gobierno de colombia, 2018; Rodríguez-García et al., 2017). Concurrently, the incidence and prevalence of both entities appears to be increased among the aging population. Likewise, among patients with cancer, cerebrovascular disease is the second most common neurological manifestation following metastases (Zhang et al., 2006). Yet, this association is often disregarded in clinical practice. Stroke can occur at any point during malignancy and it can even be the first manifestation of an occult malignancy in up to 3% of patients (Uemura et al., 2010). Furthermore, autopsy findings of cancer patients reveal stroke in 15 % of cases; half of which are asymptomatic (Kim et al., 2010). Given that stroke can be a potential first sign of neoplasia (Uemura et al., 2010), it demands an accurate etiological diagnosis in order to gear therapy accordingly and improve clinical outcomes (Uemura et al., 2010; Kim et al., 2010). Prognosis, disability and health expenses are greater in patients with cancer and stroke compared with subjects without cancer (Dearborn et al., 2014). Therefore, it is important to search for occult malignancy in acute stroke patients (Uemura et al., 2010). The aim of the present review is to describe the clinical characteristics, risk factors, biomarkers and treatment approaches in patients with cryptogenic stroke associated with neoplasia. Additionally, we will provide physicians with some additional clues for suspecting occult malignancy as the potential silent cause of cerebrovascular disease.
2. What are the possible scenarios in which cancer can be associated with ischemic stroke?
Currently, there is no consensus on how to identify cancer risk in acute stroke patients.
In the clinical practice, there are four situations in which cerebral ischemic disease could be associated with neoplasia: i. Subjects with recent cancer diagnosis who present with a stroke of unknown mechanism (cryptogenic stroke with active cancer), ii. A known cancer patient with a typical stroke etiology; iii. A stroke in a patient who had cancer but has now recovered (cryptogenic stroke with inactive cancer) and iv. A patient with an occult malignancy that manifests with a stroke; (cryptogenic stroke with unknown neoplasia) (Kneihsl et al., 2016). The first and second groups have been cataloged as the active cancer group. This represents a common clinical scenario. Usually these patients have been recently diagnosed (within the last 6–12 months) and underwent any type of cancer treatment and may or may not have local or distant recurrences (Kim et al., 2012; Lee et al., 2014). The patients in the inactive cancer group (group three); represent disease survivors. Commonly, their time since diagnosis is above 12 months (Kassubek et al., 2017; Guo et al., 2014). Finally, and probably the most frightening and challenging group corresponds to the cryptogenic stroke patients whose cancer is yet to be uncovered (Selvik et al., 2015). As the case depicted on Fig. 1 In this situation, the clinician’s high level of suspicion and expertise drives the subsequent clinical conduct. Therefore, this latter group represents a real challenge, as it is not necessary to screen for cancer in every case of cryptogenic stroke since it isn’t cost effective (Selvik et al., 2015). Consequently, it is primordial to limit the scenarios in which cancer should be considered as part of the differential diagnosis in patients with stroke of unknown etiology.
Fig. 1. Typical brain MRI of cancer-associated acute ischemic stroke.
A 60-year-old man who initially presented with left hemiparesis. Subsequently acute ischemic stroke was diagnosed. The brain DWI-MRI showed multiple infarct lesions in multiple vascular territories on bilateral hemispheres (A, B, C) and small-scattered lesions in the right and predominant in left hemisphere (D). At the same time, a chest-CT showed a pulmonary nodule and pulmonary embolism (E), after work-up a pulmonary adenocarcinoma was diagnosed 1 month later. The PET-CT (F) documented the right apical active nodule and mediastinum multiple ganglia.
3. What is the relationship between cryptogenic stroke patients and cancer
A stroke of cryptogenic etiology is an ischemic stroke with no identified cause despite an exhaustive investigation (Dearborn et al., 2014; Saver, 2016). The underlying mechanisms are varied and categorized as embolic and non-embolic. Common causes of embolic stroke are cancer, occult paroxysmal atrial fibrillation, among others (Fig. 2) (Fonseca and Ferro, 2015). In the general population, the frequency of cryptogenic stroke is 20–40 % (Fonseca and Ferro, 2015; Bayona-Ortiz et al., 2017). However, an etiology is not found in up to 40–51 % of patients with cancer (Quintas et al., 2018; Gon et al., 2016; Navi et al., 2014). Strikingly, approximately 20 % of patients with stroke of undetermined cause could have an occult malignancy at the time of presentation (Selvik et al., 2018). Simultaneously, stroke has been described as the first manifestation of an unknown neoplasia in up to 3% of patients (Cocho et al., 2015). Therefore, patients who present with a cryptogenic stroke are at increased risk of having an occult malignancy. The pathogenesis of stroke seems to be different in subjects without neoplasia compared with cancer patients, but the evidence is controversial (Cocho et al., 2015; Grisold et al., 2009). Even though classic etiologies of stroke such as large artery disease and cardioembolic source are frequent among patients with malignancy, cryptogenic stroke is more frequent and has a stronger association with cancer (Dearborn et al., 2014; Cocho et al., 2015).
Fig. 2. Pathophysiology of cancer-associated thrombosis remains partially unknown. This figure illustrates the diagnostic sequence of a patient with stroke considering cancer as a possible etiology after ruling out other nosological entities. Finally, the pathogenesis and risk factors of cancer-associated thrombosis are presented into three categories: patient characteristics, treatment-related issues, and cancer-specific factors.
4. What kind of cancer associated risk factors play a role in stroke development?
Both cancer and cerebrovascular disease share a significant amount of risk factors. These are more common in the aging population and are burdened with vascular risk factors. Indeed, reports have showed that the prevalence of such vascular risk factors (hypertension & smoking, hyperlipidemia, diabetes mellitus, alcoholism, obesity, atrial fibrillation) is similar between cancer stroke patients and non-cancer stroke patients (Dearborn et al., 2014; Quintas et al., 2018; Selvik et al., 2014). Given the high prevalence and pathogenic effect of vascular risk factors, it is not surprising that these are still the most frequent cause of stroke, even among cancer population (Dearborn et al., 2014). On the same note, reports have demonstrated that the proportion of conventional stroke mechanisms (atherosclerotic, cardioembolic, lacunar) are approximately equal between patients with and without cancer (Dearborn et al., 2014). Additionally, some studies have demonstrated that atherosclerosis is the most common cause of ischemic stroke in patients with neoplasia (Kim and Lee, 2014). However, data is conflicting as other studies have established that on the contrary, conventional vascular risk factors were less relevant in ischemic stroke cancer patients (Shin et al., 2016).
The mechanisms of stroke in the context of cancer is not entirely elucidated. Since vascular risk factors are highly prevalent on stroke patients regardless of their cancer status, whether both diseases processes arise independently and simultaneously or if cancer has a direct influence on the pathophysiology of stroke is still unclear. (Fig. 3).
Fig. 3. Interaction of multiple factors involved in the pathogenesis of arterial thrombosis in cancer patients. IMIDs: immunomodulatory drugs; TKI: tyrosine kinase inhibitors; VEGF: vascular endothelial growth factor; TNF: tumor necrosis factor; u-PA: urokinase-type plasminogen activator; t-PA: tissue plasminogen activator; PAI-1 and 2: plasminogen activator inhibitor-1 and 2; TF: tissue factor; CVCs: central venous catheter; BCR/ABL: Philadelphia chromosome.
4.1. Coagulopathy
Hypercoagulability is regarded as the most significant mechanism of cryptogenic stroke in patients with cancer (Grazioli et al., 2018). It was first described by Trousseau in 1865 in the setting of gastric carcinoma and migratory thrombophlebitis. Tumor cells release pro-coagulant molecules, tissue factor and cancer procoagulant (a cysteine protease), that heightens the coagulation cascade. In addition, other cytokines are released such as TNF-alpha, IL-1 and IL-6 (Grisold et al., 2009). These molecules act as pro coagulants by: i). Inducing cells to express tissue factor, ii). Inhibiting Protein C activation and iii). Shedding vascular endothelial cells and therefore further thickening blood (Dearborn et al., 2014). It constitutes a paraneoplastic and yet poorly understood phenomenon that decreases survival in affected individuals (Schwarzbach et al., 2012; Lee et al., 2017).
Other coagulopathies, including disseminated intravascular coagulation (DIC) present more frequently in stroke patients with cancer (Dearborn et al., 2014). Several studies have tried to use laboratory markers to quantify coagulopathy. d-dimer is a marker of an activated coagulation system. Cancer stroke patients have higher d-dimer levels compared to patients with stroke and no cancer (Kim et al., 2010; Dearborn et al., 2014; Quintas et al., 2018; Schwarzbach et al., 2012; Lee et al., 2017). It is also an independent predictor for stroke of non-conventional mechanisms and is significantly associated to cancer in multiple studies (Kim et al., 2010; Álvarez-Pérez et al., 2012). Seok et al. found a higher prevalence of micro embolisms in transcranial doppler recordings of cancer stroke patients, predominantly in those with unconventional stroke mechanisms which correlated significantly with d-dimer levels (Seok et al., 2010). However, d-dimer is a non-specific marker, it can become elevated in numerous circumstances including cancer patients without stroke (Schwarzbach et al., 2012).
4.2. Cancer site and histologic subtype
Adenocarcinoma of the lung and adenocarcinomas of the gastrointestinal tract are the most common type of malignancies among cancer stroke patients across multiple cohorts (Kim et al., 2010; Dearborn et al., 2014; Navi et al., 2014). Adenocarcinomas are the most common histologic subtypes in stroke and cancer series (Dearborn et al., 2014; Quintas et al., 2018; Lee et al., 2017; Álvarez-Pérez et al., 2012). This is probably because they are frequently associated with clotting disorders via its production and secretion of mucin, a high molecular weight particle that interacts with cell adhesion molecules (P and l-selectins) and induces micro thrombi formation (Schwarzbach et al., 2012). Other common cancers in stroke cohorts are prostate, breast, bladder, gynecological cancer, pancreatic and melanoma (Dearborn et al., 2014; Quintas et al., 2018; Zhang et al., 2007; Navi et al., 2015). Hematological malignancies like non-Hodgkin lymphoma have also been reported (Quintas et al., 2018). Outstandingly, patients with smoking-related cancers have higher risk of stroke (lung, colon, bladder, rectum, pancreas, kidney, stomach, and head and neck) (Andersen and Olsen, 2018).
Additional but infrequent direct cancer mechanisms for stroke also include the occurrence of an embolism to the brain from heart tumors, hematologic malignancies like polycythemia vera’s hyperviscosity syndrome and direct infiltration of vascular structures such as the case of intravascular lymphoma (Dearborn et al., 2014; Grisold et al., 2009).
4.3. Non-bacterial thrombotic endocarditis (NBTE)
In NBTE, sterile vegetations in the cardiac valves that are thought to develop due to valve attachment of disrupted fibrin that forms a matrix for platelets to bind. One of the most common targets for emboli due to NBTE is the cerebral circulation. NBTE is found as one of the most prevalent risk factors in cancer and stroke in studies (Navi et al., 2014; Sun et al., 2016). It is related with mucinous carcinomas mainly of pancreatic origin (Grisold et al., 2009).
4.4. Tumor mass effect
The tumor mass itself or its surrounding edema can cause direct compression of blood vessels in the brain, causing ischemia of the affected territory. This must be differenced from a hemorrhagic conversion of a brain metastasis leading instead to a hemorrhagic stroke. This phenomenon has also been described in primary brain neoplasia such as high-grade glioma and benign tumors like meningioma. Surgery of this type of tumors is related with perioperative stroke but the mechanism is not defined yet (Grisold et al., 2009).
No comments:
Post a Comment