Commentary (DeAngelis): Carcinomatous Meningitis: It Does Not Have to Be a Death Sentence

Leptomeningeal metastases are a growing problem, particularly among patients whose solid tumors have been controlled with increasingly effective systemic therapies. The central nervous system (CNS) can represent a sanctuary where disease can persist even if it is eradicated or controlled systemically. This is due to the blood-brain barrier that prevents most water-soluble chemotherapeutic agents from reaching tumor cells sequestered in the CNS.

In addition, aggressive treatment of brain metastases may inadvertently lead to an increased incidence of leptomeningeal metastases by prolonging survival and potentially disseminating tumor cells into the subarachnoid space during surgical resection. This is particularly true when metastases are resected from the posterior fossa, where the development of subsequent leptomeningeal tumor can be as high as 38%.[1,2] Consequently, leptomeningeal metastases are being seen with increased frequency, and they represent one of the most difficult metastatic complications of cancer to treat.

Diagnosing Leptomeningeal Metastases

Dr. Sagar provides a nice overview of leptomeningeal metastases. He outlines the common clinical picture of multilevel neurologic symptoms and signs. In particular, more signs are frequently found on neurologic examination than anticipated from the relatively few symptoms the patient reports. Diagnosis has traditionally required demonstration of malignant cells in the cerebrospinal fluid (CSF). However, modern neuroimaging with gadolinium-enhanced magnetic resonance imaging (MRI) frequently reveals the presence of leptomeningeal tumor even in the absence of a positive CSF cytologic examination. When imaging is definitive, treatment can be instituted on that basis alone.

However, normal neuroimaging in a patient with typical symptoms and signs of leptomeningeal tumor does not negate the presence of subarachnoid metastases, particularly in leukemia and lymphoma patients, almost half of whom have normal MRIs.[3] All neuroimaging should be performed prior to lumbar puncture because this invasive procedure can produce patchy dural enhancement that may be misinterpreted as leptomeningeal metastases.[4,5] In addition, complete imaging of the neuraxis is necessary for all patients, to identify bulky disease that may require focal radiotherapy and to ensure that there is no obvious mechanical obstruction to CSF flow.

Challenges of Treatment

Treatment of leptomeningeal metastases is particularly challenging. Tumor resides in the subarachnoid space, an area that most drugs cannot penetrate. Tumor cells frequently invade the cranial nerves, spinal roots, cauda equina, and even the surface of the brain or spinal cord. This further protects the disease from most chemotherapeutic agents. For many years, drugs were delivered directly into the subarachnoid space as a means of circumventing these problems. However, intrathecal drugs do not penetrate into large tumor nodules or neural tissue. In addition, the few agents (methotrexate, cytarabine, and thiotepa [Thioplex]) that can be administered safely into the CSF are not active against the majority of solid tumors-such as lung and breast cancer, and melanoma-that metastasize to the subarachnoid space.

Other Management Options

Ventriculoperitoneal shunting is often required to manage the increased intracranial pressure caused by leptomeningeal metastases.[6] Dr. Sagar advocates the use of on/off valves to enable administration of intra-Ommaya chemotherapy, but in our experience, we find these valves rarely work properly. Some patients cannot tolerate having the valve turned off to administer intrathecal chemotherapy. More importantly, with the valve turned off, the CSF does not circulate properly, and the drug remains primarily within the ventricular system without circulating appropriately throughout the entire subarachnoid space. This leads to prolonged drug exposure in the ventricles, thus increasing the risk of neurotoxicity among long-term survivors.[7] The risk of peritoneal seeding is minimal, and this concern should never be a reason for withholding placement of a shunt in patients with elevated intracranial pressure.

Radiotherapy represents the most effective strategy for reversing neurologic deficits, palliating pain, and preventing neurologic deterioration.[8,9] It can be delivered concurrently with intrathecal chemotherapy, except when it is being administered to the brain. Symptomatic sites and bulky disease should be treated with focal radiotherapy. However, the target field should not be limited to disease visualized on neuroimaging studies. For example, patients with severe lumbar radicular pain and leg weakness should receive radiotherapy to the lumbosacral spine even if neuroimaging in this area does not reveal obvious tumor nodules.

Systemic chemotherapy is being used increasingly in the treatment of leptomeningeal metastases.[10,11] There are data suggesting it is at least as effective, if not more effective, than intrathecal chemotherapy. It has the advantage of reaching tumor that has penetrated neural structures as well as nodular or bulky disease within the subarachnoid space via the vasculature. The range of available systemic chemotherapy agents is much greater and, theoretically, includes agents that are active against the primary.

However, a fundamental therapeutic problem is that leptomeningeal metastases develop late in the course of a malignancy and often from a highly resistant subclone of the primary tumor. This may be a much more intrinsic problem in the effective treatment of this disease than some of the mechanical and physiologic limitations associated with reaching tumor in this compartment. This will require the development of new and more effective agents, and perhaps, the incorporation of preventative therapies as part of initial treatment aimed at reaching the CNS and eradicating any microscopic cancer cells residing there.

References:

1. DeAngelis LM, Mandell LR, Thaler, HT, et al: The role of postoperativeradiotherapy after resection of single brain metastases. Neurosurgery24:798-805, 1989.

2. Mirimanoff RO, Choi NC: The risk of intradural spinal metastases inpatients with brain metastases from bronchogenic carcinomas. Int J Radiat OncolBiol Phys 12:2131-2136, 1986.

3. Freilich RJ, Krol G, DeAngelis LM: Neuroimaging and cerebrospinal fluidcytology in the diagnosis of leptomeningeal metastasis. Ann Neurol 38:51-57,1995.

4. Brightbill TC, Goodwin RS, Ford RG: Magnetic resonance imaging ofintracranial hypotension syndrome with pathophysiological correlation. Headache40(4):292-299, 2000.

5. Bakshi R, Mechtler LL, Kamran S, et al: MRI findings in lumbar punctureheadache syndrome: Abnormal dural-meningeal and dural venous sinus enhancement.Clin Imaging 23(2):73-76, 1999.

6. Lokich J, Levine H, Nasser I: Malignancy-related hydrocephalus: Clinicalfeatures and results of ventricular peritoneal shunt procedure in threepatients. Am J Clin Oncol 21(4):366-368, 1998.

7. Mason WP, Yeh SDJ, DeAngelis LM: ¹¹¹Indium-diethylenetriamine pentaaceticacid cerebrospinal fluid flow studies predict distribution of intrathecallyadministered chemotherapy and outcome in patients with leptomeningealmetastases. Neurology 50:438-443, 1998.

8. Gray JR, Wallner KE: Reversal of cranial nerve dysfunction with radiationtherapy in adults with lymphoma and leukemia. Int J Radiat Oncol Biol Phys19(2):439-444, 1990.

9. Chamberlain MC: New approaches to and current treatment of leptomeningealmetastases. Curr Opin Neurol 7(6):492-500, 1994.

10. Glantz MJ, Cole BF, Recht L, et al: High-dose intravenous methotrexatefor patients with nonleukemic leptomeningeal cancer: Is intrathecal chemotherapynecessary? J Clin Oncol 16:1561-1567, 1998.

11. Boogerd W, Dorresteijn LD, vanDer Sande JJ, et al: Response ofleptomeningeal metastases from breast cancer to hormonal therapy. Neurology55(1):117-119, 2000.