Blood Brain Barrier Disruption (BBBD) Chemotherapy

CMS NCA document | source_status=Closed | review_type=New | public_comment_open=False | document_id=CAG-00333N

CMS has determined that there is sufficient evidence to conclude that the use of osmotic blood brain barrier disruption (BBBD) used as part of a treatment regimen for brain tumors in Medicare beneficiaries is not reasonable and necessary under Section 1862(a)(1)(A) of the Social Security Act. Accordingly, we are issuing a national coverage determination (NCD) that states: The use of osmotic blood brain barrier disruption is not reasonable and necessary when it is used as part of a treatment regimen for brain tumors. This NCD does not alter in any manner the coverage of anticancer chemotherapy.

Decision Memo: To: Administrative File: CAG #00333N Blood Brain Barrier Disruption (BBBD) From: Steve Phurrough, MD, MPA Director, Coverage and Analysis Group Louis Jacques, MD Division Director LCDR Tara Turner, PharmD Lead Analyst James Rollins, MD, MSHA, PhD Lead Medical Officer Subject: Final Coverage Decision Memorandum for Osmotic Blood Brain Barrier Disruption (BBBD) when used as part of a treatment regimen for brain tumors Date: March 20, 2007 I. Decision CMS has determined that there is sufficient evidence to conclude that the use of osmotic blood brain barrier disruption (BBBD) used as part of a treatment regimen for brain tumors in Medicare beneficiaries is not reasonable and necessary under Section 1862(a)(1)(A) of the Social Security Act. Accordingly, we are issuing a national coverage determination (NCD) that states: The use of osmotic blood brain barrier disruption is not reasonable and necessary when it is used as part of a treatment regimen for brain tumors. This NCD does not alter in any manner the coverage of anticancer chemotherapy. II. Background We are providing a descriptive summary of brain tumors, chemotherapy for brain tumors, the blood brain barrier (BBB), and methods for its disruption so that readers who are unfamiliar with this concept can better understand this memorandum. Blood brain barrier disruption The blood-brain barrier (BBB) is a physiologic barrier that protects the brain from toxic substances, including most chemotherapeutic agents. It is created by the tight junctions between endothelial cells that line the capillaries in the brain. Factors important in determining drug-entry into the brain include molecular weight as well as lipid solubility. The BBB normally prevents passage of drugs with molecular weights greater than 180 Daltons. The BBB may be partly responsible for the poor efficacy of chemotherapy for malignant primary or metastatic brain tumors. BBBD is the disruption of the tight junctions between the endothelial cells that line the capillaries in the brain, accomplished by osmotic disruption, bradykinin or irradiation (Neuwelt; Frenkel, Diehl, et al.1980; Van Vulpen, Kal, Taphoorn, El-Sharoun, 2002; Kemper, Boogerd, Thuis et al. 2004). Osmotic disruption of the BBB, the most common technique, has revealed that separation of the cerebral endothelial tight junctions can occur, which will allow transmission of higher molecular weight molecules, as well as substances which are less lipophilic, into the brain. This decision is only evaluating osmotic disruption. Any use of BBBD throughout this document and decision relates only to osmotic BBBD. Mannitol is the most commonly used BBBD agent; this infusion is usually delivered into either the internal carotid or vertebral artery, depending on the tumor’s arterial supply, via percutaneous femoral artery catheterization. The rate and duration of mannitol infusion are critical for successful barrier disruption. After barrier disruption occurs, contrast dye is infused to document the location and extent of barrier altercation. Chemotherapeutic agents are given in conjunction with barrier disruption. Repeated treatments are given monthly as needed. Serial scans of the brain are performed to monitor the progress of treatment. While the mannitol administration is central, the BBBD process includes all items and services necessary to perform the procedure, including hospitalization, monitoring, and repeated imaging procedures. Title XVIII of the Social Security Act defines coverage of drugs used in an anticancer chemotherapeutic regimen in section 1861(t)(2). Mannitol is not an anticancer chemotherapeutic drug. As noted below, the use of radiation for the treatment of brain tumors is associated with cognitive deficits. One claimed benefit of BBBD to enhance the delivery of chemotherapy is the avoidance of radiation. This could result in preservation of cognitive function, or perhaps even improvement in cognitive deficits. Proponents claim that BBBD followed by chemotherapy can improve the effectiveness of care, quality of life and survival for patients with brain tumors. Most of the published research on the use of BBBD as part of a treatment regimen for brain tumors has been based on studies involving Primary Central Nervous System Lymphoma (PCNSL). PCNSL is a high-grade B cell malignancy that presents within the neuroaxis without evidence of systemic lymphoma. The prognosis of PCNSL is poor compared to histologically similar lymphoma occurring outside the central nervous system (CNS). As noted in a study by Roman-Goldstein and associates, the incidence of central nervous system lymphoma has been increasing in both immunologically competent and immunologically compromised patients (e.g., transplant recipients or patients with AIDS) (Roman-Goldstein, Jones, Delashaw, McMenomey et al. 1998). As this disease shows increasing incidence, atypical clinical and radiological presentations will occur. Though not commonly found in the cavernous sinus region or internal auditory canal, PCNSL is included in the differential diagnosis of these regions. This tumor is relatively rare; standardized guidelines for the baseline evaluation and response assessment of PCNSL are lacking, thus complicating comparability among clinical trials. Abrey and colleagues, as well as other international groups, have formulated recommendations to outline consensus opinion regarding baseline evaluation for all patients, standardize response criteria and outcomes measures for patient enrolled in clinical trails, and to review clinical issues unique to PCNSL (Abrey, Batchelor, Ferreri, Gospodarowicz, et al. 2005). Incidence and epidemiology of brain tumors Brain tumors are the third leading cause of cancer deaths in men ages 20 to 39 and fifth leading cause in women of that age. Some authors have noted that brain tumors might be better termed “intracranial neoplasms” since some do not arise from brain tissue (DeAngelis, 2001). We recognize the reasonableness of this point and discuss the various origins of these neoplasms later in this section. Similarly, because the brain is part of the CNS these tumors are sometimes also referred to as CNS tumors or CNS neoplasms. We preferably use the term brain tumors in this decision memorandum as we believe that using this simpler terminology will facilitate the general public’s understanding of this document. However, we use the other terms when greater precision is needed. The American Cancer Society estimates that 18,820 malignant tumors of the brain or spinal cord (10,730 in men and 8,090 in women) will be diagnosed during 2006 in the United States (ACS 2006). Approximately 12,820 people (7,260 men and 5,560 women) will die from these malignant tumors (a mortality rate of 6 per 100,000). This type of cancer accounts for approximately 1.3% of all cancers and 2.2% of all cancer-related deaths. Both adults and children are included in these statistics. According to the National Cancer Institute (NCI), the incidence and mortality rates for cancers that originate in the brain and CNS has remained relatively unchanged in the last decade (SEER 2006). Both incidence and mortality rates are substantially higher for Whites than for other racial/ethnic groups. Regardless of racial/ethnic group, men have higher incidence and mortality rates than women. Brain and other CNS cancers are the second leading cause of cancer-related death in children and make up 21 percent of all childhood cancers. In comparison to adults, the absolute number of brain and CNS cancer deaths in children is smaller and survival rates are higher. According to the NCI, between 1973 and 1985, the total age-adjusted cancer incidence in the United States (all races, men and women) rose by 10.7%, with an average annual percentage change of +0.9% (Greig, Ries, Yancik, Rapoport, 1990). Analysis of the reported age-specific incidence of primary malignant brain tumors over that same period of time revealed that incidence rates increased dramatically between 1973/1974 and 1985. Compared to 1973/1974, the 1985 incidence of primary malignant brain tumors increased for persons aged 75-79, 80-84, and 85 years of age by 187%, 394%, and 501%, respectively. Similar increases were found in both men and women, analyzed separately and in combination. Average annual percentage changes in primary brain tumor incidence were +7.0%, +20.4%, and +23.4% in these age ranges, respectively. The incidence of primary brain cancer in younger persons varied little over the same period of time. Two possible causes have been hypothesized that might explain the increased incidence in the elderly: the introduction and extensive use of x-ray computed tomography since 1973 that may account for the detection of more tumors and/or a true increase in incidence occurring independently of diagnostic advances. Both environmental factors (e.g., ionizing radiation, immunosuppression, exposure to vinyl chloride, benzene and other organic compounds, heavy metals,), as well as genetic risk factors (e.g., neurofibromatosis, tuberous sclerosis, Multiple Endocrine Neoplasia type 1) have been implicated in development of brain tumors (Salvatore Weitberg, Mehta, et al. 1996; Moss, 1985; Tomlinson, 1997; Young, Povey 1998; Gutmann, Aylsworth, Carey et al. 1997). Transplant recipients and patients with the acquired immunodeficiency syndrome (AIDS) have substantially increased risks for PCNSL (Levin, Leibel, Gutin, 2001; Schabet M, 1999). Classification of brain tumors Tumors involving the brain can be categorized as primary (tumors originating in the brain), or secondary (tumors arising from other organs and metastasizing to the brain). The World Health Organization (WHO) classifies CNS tumors by their patterns of differentiation and presumed cell of origin (see Appendix C). Approximately 70% of symptomatic primary CNS tumors arise within the substance (parenchyma) of the brain and spinal cord. The remainders arise within the tissues surrounding the brain (meninges), pituitary or pineal glands. Glial cell tumors are the most common brain tumors. Examples of neuroepithelial tumors that are likely to occur in the elderly include astrocytomas, glioblastomas, ependymal tumors, and oligodendrogliomas. Recent evidence has suggested that oligodendrogliomas may be more common than previously thought; this is of importance because these tumors are chemosensitive. Glial tumors account for 50 to 60% of primary tumors, meningiomas account for 25%, schwannomas for 10%, and all other CNS tumors for the remainder. Occasionally, CNS tumors arise from cells not considered central nervous system in nature. About 1 in 4 patients with cancer will develop tumors that spread to the CNS, though some note that brain metastases outnumber primary neoplasms by at least 10 to 1, and may occur in 20% to 40% of cancer patients (Patchell 2003). These tumors include germ cell tumors (histologically identical to those of testicular or ovarian origin), PCNSL: both of which originate in the brain as well as metastatic tumors (originate from outside the brain). The most common primary cancers metastasizing to the brain include lung cancer (50%), breast cancer (15 to 20%), cancers of unknown primary sites (10 to 15%), melanoma (10%), and colon cancer (5%) (Patchell 2003; (Nelson JS, Von Deimling A, Peteren 2000), though almost any cancer has that potential. Metastatic tumors typically arise where the white and gray matter of the brain meet. The symptoms depend upon the function of the affected part of the brain; they can include headaches, seizures, or no symptoms at all, when first detected. About 15% of patients who die of cancer have symptomatic brain metastasis; an additional 5% suffer spinal cord involvement. Treatment of brain tumors The histologic type, the location of the cancer, and the general condition of the patient determine what therapy is appropriate for patients with brain tumors. The three main therapeutic strategies include surgery, radiotherapy, and chemotherapy. Other therapeutic categories under development include immunotherapy, gene therapy, and antiangiogenesis therapy. Although surgery for patients with brain or spinal cord tumors is rarely curative, it is the most important treatment for patients with accessible tumors other than PCNSL. Surgery can be used to confirm a diagnosis, relieve intra-cranial pressure, and to improve symptoms as well as control seizures. Corticosteroid administration in association with surgery can also reduce the risk of worsening function by preventing postoperative swelling. The evidence seems to indicate that more complete surgical removal of a tumor improves both quality of life as well as survival. The exception to this rule is gliomas. According to Albert and associates, if a tumor exhibits contrast enhancement before surgery, a postoperative contrast-enhanced MRI scan obtained within 3 days of resection accurately predicts the extent of residual tumor and thereby helps to establish the prognosis. Surgeons’ clinical estimates of the extent of resection are not thought to be reliable (Albert, Forsting, 1994). The second mode of treatment for patients with brain tumors is radiotherapy. This form of treatment can be provided in a number of ways, and is often used when the entire primary tumor cannot be surgically removed. External-beam radiation (whole brain radiation therapy), the traditional form of radiation therapy, delivers radiation from outside of the body. Hyperfractionation is a modified form of external-beam radiation that involves applying less intense but more frequent doses of radiation. Some benign tumors are treated with external-beam radiation to prevent recurrence, even if the entire primary tumor has been surgically removed. They also may be treated with radiation at the time of recurrence. Stereotactic (or stereotaxic) radiosurgery uses a large dose of radiation to destroy tumor tissue in the brain. High doses of radiation are directed precisely to the required areas. Most nearby tissues are not damaged by this procedure. Stereotactic radiosurgery can be done in one of three ways: a linear accelerator (high-energy photon radiation); gamma knife (cobalt 60); or heavy charged particle beams (e.g., protons and helium ions.) Stereotactic radiosurgery is most commonly used to treat small benign tumors as well as both primary and secondary brain cancers, and can be used either alone or along with whole-brain radiation therapy. Stereotactic radiotherapy uses the same approach as stereotactic radiosurgery to deliver radiation to the target tissue. However, stereotactic radiotherapy uses multiple small fractions of radiation as opposed to one large dose. Giving multiple smaller doses may improve outcomes and minimize side effects. The advantage of using targeted radiation is that the surrounding, healthy tissue is left undestroyed. It often is used in addition to external-beam radiation, especially in cases of malignant gliomas and metastases that are in deep or sensitive areas of the brain. Though postoperative radiotherapy improves the quality of life and prolongs the duration of survival with high-grade tumors (e.g., anaplastic astrocytoma or glioblastoma multiforme), its role in patients with low-grade (particularly asymptomatic) tumors is uncertain. In asymptomatic patients with low-grade astrocytomas or oligodendrogliomas, radiotherapy is often postponed until symptoms develop. Some feel that conventional radiotherapy offers modest palliation (Alexander, Moriarty, Davis, Wen, 1995). There are recommendations that brain and spinal cord gliomas should be treated with high doses of irradiation (5,500 to 6,000 cGy for brain tumors, and 4,500 to 5,000 cGy for spinal tumors). Irradiation is applied to the tumor as well as the surrounding region. Stereotactic radiosurgery can be used to treat metastasis and to “boost” conventional irradiation for gliomas, though its efficacy has been difficult to establish (Chang Adler, Hancock, 1998). One potential complication to patients being treated with radiotherapy for brain tumors is its potential effect on cognitive functions. This adverse event has been documented in children (Radcliffe J, Bunin GR, Sutton LN, 1994; Maddrey AM, Bergeron JA, Lombardo ER, et al., 2005) as well as in adults (Kramer, Crowe, Larson, et al. 1997; Laack, Brown, 2004). Meyers and Scheibel were one of the first to publish on this subject (Meyers, Scheibel, 1990). They noted that cancer patients often developed cognitive and behavioral alterations during or after radiation therapy, chemotherapy, or immunotherapy. They also note that some impairments are acute and reversible, while others persist after cessation of treatment or have a delayed onset. Gregor and associates also noted that neuropsychometric deficits are common after radiation treatment for brain tumors, and could be related to the timing of treatment, as well as the specific radiation technique (Gregor, Cull, Traynor, et al. 1996). And more recently, after noting that individuals with low-grade gliomas, PCNSL, and those undergoing prophylactic cranial irradiation for systemic malignancies often suffered neurocognitive sequelae, Byrne proposed the use of non-steroidal anti-inflammatory drugs (NSAIDs) as a means of combating this complication (Byrne 2005). The third form of treatment of brain cancer is chemotherapy. A number of chemotherapeutic drugs, as well immunotherapy agents have been used for this condition. One of the first chemotherapeutic agents used for brain cancer was methotrexate (MTX). Newer agents used to treat certain forms of brain cancer include procarbazine, platinum analogs (cisplatin, carboplatin), the nitrosoureas, BCNU, and etoposide. Temozolomide, granulocyte-macrophage colony-stimulating factor, as well as rituximab are immunotherapeutic agents currently being investigated for the treatment of brain cancers. A problem with chemotherapy is that, due to the size of the molecule, there are few chemical agents that can cross the blood-brain barrier to get to the tumor. An additional concern with chemotherapy is that these agents work by interrupting mitosis, the process of cell division. By nature, many brain tumors grow slowly, so slowing tumor growth by these chemotherapy agents does not result in significant clinical improvement. Some cancer cases are treated with chemotherapy after surgery and radiation. Chemotherapy can be used as a radio-sensitizing agent with radiation to control a recurrent tumor and to treat patients who can no longer tolerate radiation therapy. Studies have shown that some patients who receive chemotherapy for malignant tumors have improved survival rates compared to patients who do not, but the effectiveness of chemotherapy agents is limited and depends on the tumor type (Dinnes, Cave, Huang, Milne 2002; Kim, Lee, Yun, Kim, et al. 2005). III. History of Medicare Coverage Medicare is a defined benefit program. An item or service must fall within a benefit category as a prerequisite to Medicare coverage. § 1812 (Scope of Part A); § 1832 (Scope of Part B) § 1861(s) (Definition of Medical and Other Health Services). BBBD used as part of a treatment regimen for brain tumors is considered to be within the following benefit categories: inpatient hospital services (§1861 (b)), and physicians’ services (§1861 (q)). This may not be an exhaustive list of all applicable Medicare benefit categories for this item or service. Medicare does not currently have a National Coverage Determination for BBBD used as part of a treatment regimen for brain tumors. IV. Timeline of Recent Activities Date Action July 11, 2006 CMS accepted a formal request for non-coverage of BBBD used as part of a treatment regimen for brain tumors. A tracking sheet was posted on the web site and the initial 30 day public comment period commenced. August 10, 2006 The initial 30 day public comment period ended. Thirty-nine comments were received. December 27, 2006 CMS posted its Proposed Decision Memorandum and opened a 30 day public comment period. January 26, 2007 The public comment period ended. Eleven comments were received. V. FDA Status Blood brain barrier disruption as a procedure is not regulated by the FDA. The individual chemotherapeutic agents and BBB disruption agents are FDA approved drugs and/or biologics. Mannitol does not have FDA approved labeling for disruption of the BBB. Mannitol (marketed as Osmitrol and generics) has labeled indications for the promotion of diuresis, in the prevention and/or treatment of the oliguric phase of acute renal failure before irreversible renal failure becomes established; the reduction of intracranial pressure and treatment of cerebral edema by reducing brain mass; the reduction of elevated intraocular pressure when the pressure cannot be lowered by other means, and promoting the urinary excretion of toxic substances. VI. General Methodological Principles When making national coverage determinations, CMS evaluates relevant clinical evidence to determine whether or not the evidence is of sufficient quality to support a finding that an item or service falling within a benefit category is reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member. The critical appraisal of the evidence enables us to determine to what degree we are confident that: 1) the specific assessment questions can be answered conclusively; and 2) the intervention will improve health outcomes for patients. An improved health outcome is one of several considerations in determining whether an item or service is reasonable and necessary. A detailed account of the methodological principles of study design that are used to assess the relevant literature on a therapeutic or diagnostic item or service for specific conditions can be found in Appendix A. In general, features of clinical studies that improve quality and decrease bias include the selection of a clinically relevant cohort, the consistent use of a single good reference standard, and the blinding of readers of the index test, and reference test results. Public comment sometimes cites the published clinical evidence and gives CMS useful information. Public comments that give information on unpublished evidence such as the results of individual practitioners or patients are less rigorous and therefore less useful for making a coverage determination. CMS uses the initial public comments to inform its proposed decision. CMS responds in detail to the public comments on a proposed decision when issuing the final decision memorandum. VII. Evidence A. Introduction We are providing a summary of the evidence that we considered during our review. The evidence reviewed to date in this final decision memorandum includes the published medical literature on BBBD reviewed as part of our internal technology assessment as well as all articles and studies submitted during the two public comment periods. B. Discussion of evidence reviewed 1. Question: Is the evidence sufficient to conclude that blood brain barrier disruption, when used as part of a treatment regimen for brain tumors, improves patient-centered health outcomes in Medicare beneficiaries, compared to therapies that do not include blood brain barrier disruption? 2. External technology assessments CMS did not commission an external technology assessment on this issue. We are aware of a November 2001 technology assessment performed by the Institute for Clinical Systems Improvement (ICSI), entitled Blood Brain Barrier Disruption Chemotherapy. The committee noted the difficulty comparing results from different trials and that few patients with any given tumor type were studied. It called attention to the likelihood of selection bias. The committee summary included seven points, which are abstracted below: Long term effects of repeated BBBD procedures are unknown. BBBD used as part of a treatment regimen for brain tumors is acceptably safe when performed by experienced physicians in large, regional centers. Protocols using high dose methotrexate for PCNSL with or without BBBD consistently produced response rates > 75%. The use of BBBD may preclude the need for whole brain radiation therapy. Some patients with anaplastic astrocytoma or glioblastoma multiforme experience tumor shrinkage. For CNS germ cell tumors the relative efficacy compared to conventional chemotherapy is unknown. Response rates > 69% were seen for the few evaluable cases of primitive neuroectodermal tumor (PNET)/medulloblastoma. There are insufficient data on patients with metastatic disease to make a comparison to conventional chemotherapy. (Accessed 12/13/06 athttp://www.icsi.org/knowledge/detail.asp?catID=107&itemID=268) 3. Internal technology assessments CMS performed an extensive literature search utilizing PubMed for new randomized controlled trials (RCTs) and systematic reviews evaluating the use of BBBD used as part of a treatment regimen for brain tumors. The literature search was limited to the English language and specific to the human population. Literature search Due to the paucity of published information on this subject, most of the studies evaluating the use of blood brain barrier disruption in patients with brain tumors were supplied by those most familiar with the technique. This information was supplemented with additional information from MEDLINE, Cochrane Review, ECRI, NCI, as well as multiple oncology, medical and surgical textbooks. Peer-reviewed articles written in English were reviewed. Search terms included blood brain barrier disrupters, blood brain barrier opening, blood brain barrier modification, brain cancer (both primary and metastatic), primary CNS lymphoma, malignant brain tumors, and mannitol. We focused the review on original reports of the use of hyper-osmotic agents used as an adjunct to chemotherapeutic agents in the treatment of primary and metastatic brain cancer. We also reviewed original reports of the use of osmotic blood brain barrier disrupters. A review of the literature has failed to reveal any published RCT results. A number of case studies, case series and nonrandomized controlled study designs have been published on the association between BBBD and the treatment of intracranial malignancy. Evidence review Early studies on the use of blood brain barrier disrupters A number of original studies evaluated the feasibility of using osmotic agents as blood brain barrier disrupters to increase chemotherapeutic agents to the brain (Neuwelt, Frenkel, Diehl, Maravilla et al. 1980; Neuwelt EA 1980; Neuwelt, Specht, Howleson et al. 1983). Neuwelt and associates were among the first to study the reversibility of BBB disrupters in patients with malignant brain tumors (Neuwelt, Frenkel Diehl, Vu et al. 1980). In this case series, 5 subjects with gliomas or metastatic brain tumors (glioblastomas, anaplastic astrocytoma, metastatic breast cancer) were given mannitol infusions intracranially, followed by an intravenous contrast agent (technetium pertechnitate). The intervention resulted in good to excellent blood brain barrier disruption in 4 of 5 subjects. Two transient complications occurred in separate subjects (seizures and aphasia). And although a single non-transient complication occurred (a superficial wound infection at the burr hole site in 1 subject), reversible transient osmotic barrier disruption was achieved 15 times in five patients without additional toxicity. This study provided evidence that a metastatic or glioblastoma tumor could have a blood brain barrier intact to an intravenous contrast agent, which only becomes permeable to this contrast agent after osmotic disruption. In a later study, Roman-Goldstein and associates demonstrated that non-ionic iodine-based contrast medium was associated with a lower incidence of seizures when injected intravenously in conjunction with osmotic blood brain barrier disrupters, compared to ionic contrast media (Roman-Goldstein, Clunie, Stevens, et al. 1994). Based on this early study which used both primary as well as metastatic brain cancer, contrast enhanced CT is the preferred method to image disruption because it has better spatial resolution than radionuclide imaging techniques. Another early study evaluated whether or not chemotherapy drug levels following BBBD are correlated with the degree of barrier disruption measured by CT scan and radionuclide scans. Neuwelt and associates performed one of the first studies to evaluate BBBD used as part of a treatment regimen in patients with malignant brain tumors (Neuwelt, Diehl, Vu, et al. 1981). In this case series, the authors monitored intra-carotid delivery of methotrexate (MTX) in six patients with malignant glial tumors who had received mannitol as an osmotic BBB disrupter. MTX was chosen during this initial trial because of its potential for low toxicity during direct exposure, its reported response in brain tumors, and the availability of a reliable assay method. During the study, a total of 33 disruptions occurred. Two subjects showed clinical improvement, one of whom had evidence of tumor regression by CT scan. No significant or permanent adverse neurologic or systemic sequelae occurred among participants. The neuroradiologic evaluation showed that MTX in the tumor persisted longer after BBB disruption than without disruption. This study also revealed that cerebrospinal fluid MTX levels were not a sensitive measure of the degree of barrier disruption as measured by either CT or brain scan. Neuwelt and associates explored concurrent tumor regression in areas distant to barrier opening (Neuwelt, Hill, Frenkel 1984). This case series included 3 subjects, each with a distinct type of brain malignancy: metastatic breast cancer, glioblastoma, and PCNSL. They all had objective responses to combination chemotherapy in conjunction with BBB modification in those areas of the brain perfused. This was documented by serial CT studies confirming reduction in the size of the intracranial lesions, and physical examination that confirmed clinical improvement. The study also showed that osmotic BBB disruption increases drug delivery not only to the tumor but also to the surrounding brain area. Subsequent to the procedure, each patient developed the occurrence or recurrence of CNS disease in areas not directly perfused by the chemotherapeutic agent. The authors concluded that though drug resistance could partially explain treatment failures, another possible cause could be that drug delivery to the tumor could be seriously affected by a partially or completely intact BBB. This could explain tumor regression seen only in those areas of the brain undergoing BBB disruption. The use of blood brain barrier disruption as part of therapy for brain tumors A number of studies have been conducted using BBBD along with chemotherapy for the treatment of various types of brain malignancies, including primary disease. Neuwelt and associates, as well as other authors, have performed a number of case series studies evaluating the use of BBBD in patients with PCNSL of the brain (Neuwelt, Frenkel, Gumerlock, et al. 1986; Neuwelt, Goldman, Dahlborg, Crossen et al. 2000; Crossen, Goldman, Dahlborg, Neuwelt 1992; Dahlborg, Henner, Crossen, Tableman et al. 1996; Roman-Goldstein, Jones, Delashaw et al. 1998; McAllister, Doolittle Gustadisegni, Kraemer et al. 2000; Neuwelt, Goldman, Dahlborg et al. 1991; Kraemer, Fortin, Doolittle, et al. 2001; Tyson, Siegal, Doolittle, Lacy et al. 2003; Ferreri, Abrey, Blay, Borisch et al. 2003; Neuwelt, Guastadisegni, Varallyay et al. 2005; Abrey, Batchelor, Ferreri, Gospodarowicz et al. 2005). Other types of primary brain tumors that BBBD along with chemotherapy has been used for on include malignant glial tumors (Neuwelt, Diehl, Vu, et al. 1981), non-glial primary brain tumors (Dahlborg, Petrillo, Crossen, Roman-Goldstein et al. 1998); and glioblastomas (Neuwelt, Howieson, Frenkel, Specht et al. 1986; Hall, Doolittle, Daman, Bruns et al. 2005), and germinoma (Neuwelt, William, Mickey et al. 1994). BBBD as part of a treatment regimen for brain tumors has also been used on a number of cancers metastatic to the brain. Some studies have evaluated individual tumor types such as melanoma (Neuwelt, Specht, Barnett, Dahlborg et al. 1987 as well as breast cancer (Tyson, Kraemer, Hunt, Muldoon, Orbay, et al. 2006). Other studies have evaluated the use of BBBD in the setting of both primary and secondary brain cancer (Neuwelt, Dahlborg 1987; Roman-Goldstein, Clunie, Stevens, Hogan, et al. 1994; Roman-Goldstein, Mitchell, Crossen, William et al. 1995; Williams, Henner, Roman-Goldstein, Dahlborg, et al. 1995). We discuss the use of BBBD used as part of a treatment regimen for specific types of brain tumors below. The first section describes the evidence on primary intracranial malignancies, i.e. those malignancies arising in the brain whether or not the cell type is neural, glial, or other. The second section describes the evidence for metastatic malignancies, i.e. tumors that originated in other locations such as breast or skin. Primary intracranial malignancies Primary Central Nervous System Lymphoma The literature has revealed a large number of studies evaluating the use of BBBD and chemotherapeutic agents for PCNSL. Neuwelt and associates were among the first to evaluate mannitol for BBBD in conjunction with MTX as well as other chemotherapeutic agents in the treatment of PCNSL (Neuwelt, Balaban, Diehl, et al. 1983). The first case involved a 37 year old subject with PCNSL treated with MTX, leucovorin rescue, procarbazine, and cyclophosphamide. He received nine courses of treatment with the above mentioned drugs. CT studies showed a decrease in tumor size with each treatment. Further CT studies showed no evidence of tumor before his eighth treatment. At the time of the publication, this subject had complete tumor regression and was neurologically intact. The second case involved a 60 year old with PCNSL who received MTX as well as cyclophosphamide. Despite the response after six treatments, the patient decided to discontinue treatments temporarily for geographic reasons, and was subsequently treated with radiation. Though complete tumor regression occurred, the subject died 12 months after diagnosis. The final case involved a 67 year old with PCNSL. The patient initially responded to cranial irradiation to the entire brain, but a few months later the tumor recurred. The patient then underwent BBB disruption and received intravenous cyclophosphamide, procarbazine along with MTX. During the course of therapy, the patient received two courses of treatment, and though the patient continued to have some bilateral leg weakness at the time of publication of the article (almost a year after getting first BBBD treatment), there was steady improvement and no evidence of tumor seen on enhanced CT. Using a prospective series, Neuwelt and associates followed the course of 12 subjects with PCNSL to determine if a new approach to diagnosis (integrating the use of needle brain biopsy and immunochemical staining for monoclonal antibodies) and treatment (use of BBBD along with chemotherapy) was effective for this condition (Neuwelt, Frenkel, Gumerlock, Braziel et al. 1986). Baseline radiographic studies were obtained (e.g., CT scans, MRIs, radionuclide brain scans), as well as special laboratory studies (e.g., cytology, immunoassays, protein electrophoresis for immunoglobulins studies). BBBD was initiated along with combinations of cyclophosphamide, MTX, leucovorin rescue, and procarbazine. Sequential BBBD and concomitant drug therapy was repeated during the course of therapy. Subjects were followed for a median of 19 months (range, 12-48 months). Results of the study revealed that CT-guided needle biopsy contributed to the diagnosis in 6 patients, and immunochemical staining methods detected monoclonal antibodies in those tested. The study also reported an initial complete response in 75% of participants, and a 1-year survival rate of 75%. The author noted that the clinical response rate and survival rate were at least as effective as radiotherapy as a primary therapeutic modality for PCNSL. Kraemer and associates explored the relationship between total dose intensity of chemotherapy delivered by BBBD in patients with PCNSL, and its relationship with survival (Kraemer, Fortin, Doolittle, Neuwelt 2001). This study involved the use of 74 patients with PCNSL who did not have systemic lymphoma or radiation treatment prior to initiation of BBBD followed by chemotherapy. Two chemotherapy protocols were used: MTX, cyclophosphamide, procarbazine (protocol 1) vs. MTX, cyclophosphamide, etoposide, and granulocyte colony stimulating factor (protocol 2). Baseline characteristics were used as potential exploratory variables (e.g., gender, age, protocol, number of disruptions, chemotherapeutic dosage intensity etc.) During this study of 74 patients, a total of 1047 BBBD procedures were performed, and total dose intensity of chemotherapy was estimated using the number of intra-arterial infusions, or a cumulative degree of BBBD score. The study revealed that survival was significantly associated with intensity of chemotherapy, i.e. increased dose intensity results in increased survival. A number of other studies have also been performed which have evaluated BBBD used as part of a treatment regimen for brain tumors along with chemotherapy in patients with PCNSL, and have tried to determine if cognitive function was affected. Neuwelt and associates followed 2 groups of patients; 13 patients that received cranial irradiation 1 to 9 months before referral (group 1), and 17 patients who received initial BBBD and chemotherapy with subsequent radiation only for tumor progression or recurrence (group 2) (Neuwelt, Goldman, Dahlborg, Crossen, et al. 2000). For patients in group 2, mannitol was used for BBBD, in conjunction with cyclophosphamide, MTX, leucovorin rescue, and procarbazine as chemotherapeutic agents. A battery of neuropsychological test was used to assess cognitive function (e.g., Wechsler Memory Scale and Revision-WMS and WMS-R; Wechsler Adult Intelligence Scale-Revised-WMS-R; Trail Making Test: Parts A and B-TMT, Karnofsky performance scores etc.), and characteristics of subjects were compared to 208 PCNSL patients abstracted from 15 published series (historical control). The results of the study revealed that the median survival for group 1 (cranial irradiation) was 17.8 months, comparable with the 20 month median survival of the historical control series. The median survival for group 2 (BBBD followed by chemotherapy) was 44.5 months. The authors also noted that improved survival was associated with preservation of cognitive function in six of seven non-irradiated complete responders observed over a 7 year period, while for those that received irradiation, several patients maintained average test results. Crossen and associates also explored whether or not cognitive function is affected in patients with PCNSL after receiving BBBD and chemotherapy (MTX, Cytoxan, procarbazine) (Crossen, Goldman, Dahlborg, Neuwalt, 1992). This study followed for 7 years 8 consecutive patients with PCNSL who received BBBD and chemotherapy. Baseline neuropsychological testing (e.g., WAIS-R, WMS-R, CFT, VLT, TMT-B) as well as Karnofsky Performance Scores (KPS) were obtained. Results of the study revealed that 7 of the 8 participants had full-scale intelligence quotient scores which tended to remain stable, as did learning performance, memory scores, and other neurobehavioral variables. Trends of summary neuropsychological test indices were stable or improved for this group. Only one participant had lower test scores compared to baseline scores that was greater than one standard deviation on 3 variables. Dahlborg and associates studied patients with and without antecedent cranial irradiation to determine if cognitive function is affected (Dahlborg, Henner, Crossen, Tableman, Petrillo et al. 1996). Fifty eight consecutive patients with PCNSL were subdivided into 2 groups: group 1-those referred to medical center at tumor progression or recurrence and after initial cranial radiation (n=19), and group 2 - those referred after initial diagnosis, not receiving cranial irradiation (n=39). Subjects ages’ ranged from 5 to 71, with 34% over age 60. Extensive clinical, neurological, ophthalmologic and neuropsychological testing was performed as well as radiological testing and KPS. Baseline demographic characteristics, as well as serial neuropsychological evaluation were also performed. Characteristics of a group of historical controls were extracted from the medical literature. Mannitol was used as the BBBD and MTX, leucovorin rescue, cyclophosphamide, and procarbazine were used as chemotherapeutic agents. The study revealed that the median survival from date of first BBBD for group 1 patients was 8.5 months, and for the group 2 patients 40 months, but with the small sample size the difference did not reach statistical significance (p<0.06). In the neuropsychological evaluation (patients were followed for 7 years), none of the patients who received only chemotherapy with BBBD and who did not receive radiation therapy suffered significant global decline in neuropsychological test results. Three of eight patients who received cranial radiation suffered declines in neuropsychological testing. This study demonstrated that for patients with PCNSL, receiving BBBD and chemotherapy preserved or improved cognitive function can be achieved, compared to PCNSL patients treated with cranial irradiation. Exploring the association between cognitive outcomes and the use of BBBD followed by chemotherapy, McAllister and associates followed a cohort of PCNSL patients after treatment (McAllister, Doolittle, Guastadisegni, Kraemer, Lacy, et al. 2000). The study consisted of 74 patients with PCNSL who had no systemic lymphoma or who had not received cranial irradiation, and who had undergone the first BBBD therapy at least 6 months prior to this study. During this study, there were 2 BBBD-enhanced chemotherapy protocols used: MTX, etoposide or cyclophosphamide, procarbazine, leucovorin rescue (protocol 1) or MTX, etoposide, cyclophosphamide, granulocyte colony stimulating hormone, leucovorin rescue (protocol 2). A battery of neuropsychological testing was performed at baseline and follow up studies were later performed (e.g., FSIQ, GMI, DRI, TMTB, etc). Other demographic characteristics and KPS were obtained at baseline. The results of the study revealed that the estimated 5-year survival rate was 42% for this group, and the median survival time was 40.7 months. Complete remission occurred in 48 patients (65%), and 36 patients continued to show complete remission response after 1 year of BBBD used as part of a treatment regimen for brain tumors followed by chemotherapy. Of these 36 patients, none demonstrated any evidence of cognitive loss. Studies have also been done to determine responses for patients with relapsed PCNSL treated with second-line BBBD followed by chemotherapy. Tyson and associates followed 37 relapsed patients with PCNSL previously treated with first-line therapy of MTX-based chemotherapy (Tyson, Siegal, Doolittle, Lacy, Kraemer et al. 2003). Patients ranged in age from 22 to 77 (mean age 57.5); all (except 1) were treated within 8 months after relapse, and 9 subjects had had previous radiotherapy. BBBD followed by chemotherapy consisted of mannitol as the BBBD, and carboplatin, etoposide, cyclophosphamide (either alone or in combination) were used as chemotherapy agents. Definitions of disease progression as well as survival were provided. Neuropsychological testing was performed at baseline as well as at completion of the study if subjects had complete response at 1 year after starting carboplatin, and patient characteristics were noted (e.g., gender, age, KPS, radiographic tumor response, survival). The results of the study revealed that the median time for survival after BBBD followed by chemotherapy was 6.8 months; however, 18% of patients survived ≥ 27 months, 24% had complete radiographic response, 11% had partial radiographic response, 32% had stable disease, and 27% had progressive disease. The median time to failure for patients with complete response and partial response was 9.1 months. A neuropsychological evaluation was performed on 4 of 8 subjects with complete response. Of these patients, there were no neurocognitive alterations in one patient, and a significant improvement in another patient who was diagnosed 11 years prior and was disease-free at the end of the study. Of the other two, one developed a systemic disease and was too ill to perform the post BBBD testing, and the other was in a stupor prior to treatment with BBBD, but completed post BBBD neuropsychological testing. No further information was given about the results. The authors concluded that for patients with relapsing PCNSL, intra-arterial chemotherapy with BBBD is a potential treatment alternative. In a separate study, Neuwalt and associates were again able to demonstrate, that for PCNSL patients receiving enhanced chemotherapy, neither enhanced chemotherapy delivery nor changes on MRI imaging following therapy were associated with decreases in cognitive function (Neuwelt, Guastadisegni, Varallyay, Doolittle 2005). Glioblastoma Neuwelt and colleagues also evaluated the use of BBBD when used as part of a treatment regimen for brain tumors in patients with glioblastoma (Neuwelt, Howieson, Frenkel, Specht et al. 1986). In this 3-arm study, 38 patients with glioblastoma (experimental group) previously treated with surgery and cranial radiation were compared to 2 historical control groups of patients with glioblastoma: one group had 14 patients treated with surgery and radiation (group 1), and the second group consisted of 8 patients with surgery, radiation, and systemic chemotherapy (group 2). Functional performance status based on KPS was measured at baseline for participants. Subjects in the experimental group received mannitol for BBBD, and cyclophosphamide, MTX, procarbazine and leucovorin rescue as chemotherapy agents. Cox Proportional Hazard was used in determining survival time, the primary outcome of interest. Risk factor effects of age, functional status, treatment, and tumor necrosis upon expected survival time were also examined. The study demonstrated an inverse relationship between age and survival time and a positive correlation between functional status and survival time (i.e. younger more functional patients given BBBD followed by chemotherapy had significantly prolonged survival compared to older patients). No significant effects upon survival time in the 3 groups were demonstrated for tumor necrosis. The median survival was 12.8 months for the group 1 controls, and 11.4 months for the group 2 controls, and 17.5 months for the experimental. This survival advantage was associated with a median KPS of 65% for those patients surviving 24 months. Neurologic as well as non-neurologic complications were reported for the experimental group (but not reported for the historical controls). Malignant glial tumors Neuwelt and associates evaluated the use of BBBD when used as part of a treatment regimen for brain tumors in patients with malignant glial tumors (Neuwelt, Diehl, Vu, et al. 1981). In this study, 6 of 8 participants had a malignant glial tumor, the other 2 patients had metastatic tumor in the brain. All subjects were treated with BBBD/MTX. The authors note that 6 patients who received BBBD followed by chemotherapy had a total of 33 disruptions. Also noted by the authors is the fact that 2 patients showed clinical improvement, one of whom had evidence of tumor regression by CT scan. Non-glial primary brain tumors Dahlborg and associates also evaluated the use of BBBD when used as part of a treatment regimen for brain tumors in patients with non-glial primary tumors (Dahlborg, Petrillo. Crossen, Roman-Goldstein et al. 1998). Thirty-four patients with histologically confirmed germ cell tumor (n=9), PCNSL (n=9), or primitive neuroectodermal tumor (n=16) were included in the study. Participant’s ages ranged from 1 to 30. Prior treatments included surgery and chemotherapy. Baseline neuropsychological testing was performed, as well as clinical evaluation. Two combination chemotherapies were used: MTX, cyclophosphamide, procarbazine, and etoposide (protocol 1), or carboplatin, etoposide, and cyclophosphamide (protocol 2). During the study, 645 BBBD and chemotherapy sessions were performed and no mortalities occurred. After treatment, of the 34 subjects included in this study, 82% had an objective response to treatment (62% with complete response, 20% with partial response). Ototoxicity was a common complication noted in patients using protocol 2 (62%). The authors note that for most patients, cognitive functioning was maintained or improved at follow up, but also notes that sample size of groups of patients with different radiation status were too small for statistical comparison. Pontine glioma Hall and colleagues evaluated the use of BBBD when used as part of a treatment regimen for brain tumors in patients with diffuse pontine gliomas (Hall, Doolittle, Daman, Bruns, et al. 2005). This study involved 8 patients with diffuse pontine gliomas, ranging in age from 2 to 44. All patients had at least 2 cycles of BBBD followed by chemotherapy and 19 was the maximum number of cycles. Chemotherapeutic agents used in the study included MTX, cyclophosphamide, etoposide, or carboplatin, cyclophosphamide and etoposide. After treatment, MR imaging revealed partial response in 2 patients, stable disease in 5, and progression of di