Most common sites on MRI of intracranial neoplastic leptomeningeal disease

Journal of Clinical Neuroscience xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn

Case study

Most common sites on MRI of intracranial neoplastic leptomeningeal disease J. Matthew Debnam a,⇑, Rory R. Mayer b, T. Linda Chi a, Leena Ketonen a, Jeffrey S. Weinberg c, Wei Wei d, Morris D. Groves e, Nandita Guha-Thakurta a a

Department of Diagnostic Radiology, Section of Neuroradiology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Department of Neurosurgery, Baylor College of Medicine, Houston, TX, USA Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, TX, USA d Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA e Austin Brain Tumor Center, Texas Oncology/US Oncology Research, Austin, TX, USA b c

a r t i c l e

i n f o

Article history: Received 20 April 2017 Accepted 21 July 2017 Available online xxxx Keywords: Leptomeningeal disease Neoplasia Magnetic resonance imaging Brain Cranial nerves

a b s t r a c t Neoplastic leptomeningeal disease (LMD) represents infiltration of the leptomeninges by tumor cells. Knowledge of the frequencies of locations of LMD on MRI may assist in early detection, help elucidate the process of leptomeningeal spread of cancer and understand how LMD affects the central nervous system. Our goal was to identify intracranial sites of neoplastic LMD predilection on MRI in patients with cytologically-proven LMD. The presence of FLAIR signal hyperintensity and T1-weighted post-contrast enhancement in the sulci of the supratentorial compartment and cerebellum and enhancement of the cranial nerves (CNs), basal cisterns, pituitary stalk, and ependymal surface of the lateral ventricles, as well as the presence of parenchymal metastasis were recorded. Within each imaging sequence, sites were ordered by prevalence and compared using McNemar’s test. The study included 270 patients. Positive MRI findings were present in 185/270 (68.5%) patients. FLAIR signal hyperintensity was significantly more common (p  0.003) in the cerebellum (n = 96) and occipital lobe (n = 92) relative to the other lobes. Leptomeningeal enhancement was also significantly more common (p  0.009) in the cerebellum (n = 82) and occipital lobe (n = 67) relative to the other lobes. Enhancement was most commonly found involving CN VII/VIII and the ependymal surface of the lateral ventricles compared to other sites. Parenchymal metastases were present in 110 (40.1%) of the patients. In conclusion, neoplastic LMD predominantly involves the cerebellum and occipital lobes, CN VII/VIII, and the ependymal lining of the lateral ventricles. Parenchymal metastases are frequently present in patients with neoplastic LMD. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction Neoplastic leptomeningeal disease (LMD) represents infiltration of the leptomeninges by tumor cells. Approximately 5% of patients with solid tumors have LMD. [1,2] Undiagnosed LMD may occur in up to 20% of patients with cancer as reported in autopsy series [3– 9]. The incidence of LMD appears to be increasing, and this increase is likely from an ascertainment bias related to improved diagnostic imaging techniques and improved therapies that have prolonged survival in the general population of patients with cancer, leading Abbreviations: LMD, leptomeningeal disease; CN, cranial nerve.

⇑ Corresponding author at: Department of Diagnostic Imaging, Section of Neuroradiology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Unit 1482, Houston, TX 77030, USA. E-mail address: [email protected] (J.M. Debnam).

to more opportunities to develop metastases such as LMD [10]. The failure of current chemotherapeutic agents to cross the blood-brain and blood-CSF barriers may also account for this rise of LMD [10,11]. The diagnosis of LMD requires one of the following three National Comprehensive Cancer Network criteria: (1) cytologic findings demonstrating tumor cells in the CSF, (2) radiologic findings of LMD irrespective of clinical findings, or (3) clinical examination findings consistent with LMD and abnormal laboratory findings in the CSF (low glucose level and elevated white blood cell and protein counts) in a patient with a cancer [12]. CSF cytologic analysis is at best only about 80–95% sensitive for the diagnosis of LMD; the failure to detect existing LMD this way may be related to the adherence of tumor cells to the leptomeninges [9,13,14]. When cytologic analysis does not show LMD, LMD can be diagnosed by neuroimaging alone in the appro-

http://dx.doi.org/10.1016/j.jocn.2017.07.020 0967-5868/Ó 2017 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Debnam JM et al. Most common sites on MRI of intracranial neoplastic leptomeningeal disease. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.07.020

2

J.M. Debnam et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx

priate clinical setting [15]. Treatment of LMD often entails intrathecal chemotherapy administered through repeat lumbar punctures or through an Ommaya reservoir, placed directly into the intraventricular system [16,17]. Knowledge of the frequencies of locations of LMD intracranially may assist in the early detection of LMD, elucidate the process of leptomeningeal spread of cancer and how LMD affects the central nervous system, potentially laying the groundwork for future studies of response to treatment of cancer. However, to the best of our knowledge, no studies have characterized the frequencies of locations of enhancement in the brain on MRI in patients with cytologically-proven LMD. Therefore, we sought to identify the most common locations of FLAIR signal hyperintensity and MRI enhancement in the intracranial compartment in patients with cytologically-proven LMD awaiting Ommaya placement at our institution, a tertiary cancer referral center. 2. Materials and methods The Institutional Review Board approved this study and waived the requirement for informed consent. Data were acquired in compliance with all applicable Health Insurance Portability and Accountability Act regulations. We retrospectively reviewed a neurosurgical database comprising 937 patients who had a clinical or radiologic diagnosis of neoplastic LMD and underwent placement of an Ommaya catheter for treatment of their disease. From this list a search was performed to identify patients in whom a brain MRI with contrast was performed and positive CSF cytologic findings diagnosing LMD either via an Ommaya or lumbar puncture was present. The MRI and collected CSF needed to be obtained within 30 days of one another. The MRI brain protocol included axial FLAIR imaging and T1weighted images at in three orthogonal orientations obtained following the intravenous administration of gadopentetate dimeglumine (Magnevist; Schering AG, Berlin, Germany). Four CAQ-certified neuroradiologists with a minimum of 15 years’ experience interpreting brain MRI agreed upon the imaging criteria to be evaluated. These included the following: the presence or absence of FLAIR signal hyperintensity and enhancement in the cerebellum (including the vermis) and involving the occipital, parietal, temporal, and frontal lobes; enhancement of the cranial nerves (CNs), including CN III, CN V, Meckel’s cave, and CN VII/VIII; enhancement involving the ependymal lining of the lateral ventricles, the basal cisterns, and the pituitary stalk; and the presence of parenchymal metastases. The MR images and reports were then reviewed by one of the four neuroradiologists. For statistical analysis, sites of FLAIR signal hyperintensity or enhancement within the left and right sides were combined, i.e. a positive finding on one side was considered a positive finding for that patient. Within each imaging sequence, sites were ordered by prevalence and compared. Prevalence of abnormalities were summarized using frequencies and percentages. As the patients had multiple sites with potential positive findings that were not mutually exclusive, their presence (yes or no) were paired at the patient level and, therefore, McNemar’s test was used to compare site prevalence. This study is a hypothesis generating study so no multiple testing adjustment was performed. A p value 0.05 was considered statistically significant. Statistical analysis was carried out using SAS version 9.4 (SAS Institute, Cary, NC). 3. Results The study included 270 patients: 143 females and 127 males aged 6–90 years (median 51.5 years, ±standard deviation 14.1 years). The most common types of malignancy were lym-

phoma (n = 84, 31.1%), breast (n = 65, 24.1%), melanoma (n = 31, 11.5%), lung (n = 30, 11.1%), leukemia (n = 15, 5.6%), glioma (n = 11, 4.1%), and other (n = 34, 12.6%). The patients underwent either 1.5 T (n = 223) or 3 T (n = 47) MRI of the brain. The MRI was performed in the time interval 30 days before, to 28 days after (median 4 days before ±10 days) the documentation of positive CSF cytologic findings. The CSF was collected via a lumbar puncture (n = 196) or from the Ommaya reservoir (n = 74). The time from the MRI to placement of the Ommaya reservoir was 105 days before to 0 days (the same day) (13 days before ±14 days) (range ±SD). A subset of patients had previously been treated with a neurosurgical procedure (n = 126, 46.7%), radiation (n = 101, 37.4%), systemic chemotherapy (n = 229, 84.8%), and intrathecal chemotherapy (n = 56, 20.7%). Two hundred-fifty of the patients were deceased at last follow-up (92.6%). Overall survival after the MRI ranged from 8 days to 9 years and 1 months (median 4 months). MRI findings consistent with neoplastic LMD involving at least one of the aforementioned categories was present in 185 of 270 patients (68.5%) (Fig. 1). FLAIR signal hyperintensity was present in 139 patients (51.5%) and was most common in the cerebellum and the occipital lobe without a significant difference (p = 0.56). Significant differences were noted in the following comparisons: occipital lobe greater than the parietal lobe (p = 0.003); parietal lobe greater than the temporal lobe (p = 0.006); and temporal lobe greater than the frontal lobe (p = 0.02). The frequencies and percentages of the locations of sulcal FLAIR signal hyperintensity ordered by prevalence with p-values based on McNemar’s test are summarized in Table 1. Leptomeningeal enhancement was present in 123 patients (45.6%) and also most common in the cerebellum which approached significance when compared to the occipital lobe (p = 0.051). Leptomeningeal enhancement was significantly more common in the occipital compared to the parietal lobe (p = 0.009). No significance difference between the remaining lobes: parietal, temporal, and frontal. These findings are summarized in Table 2. Enhancement occurred in the CNs, ependymal lining of the lateral ventricles, the basal cisterns, and the pituitary stalk in this order. A significant difference was only found between CN VII/VIII when compared to the basal cisterns (p = 0.02). These findings are summarized in Table 3. Parenchymal metastasis were present in 108 (40.1%) of the patients. In 83 of 108 (76.9%) patients, brain metastases were present prior to the development of LMD. In 25 of 108 (23.1%) patients, the brain metastases were detected at the same time as the LMD.

4. Discussion Our results demonstrate that the most common sites of both FLAIR signal hyperintensity and T1 post-contrast enhancement, representing intracranial neoplastic LMD before Ommaya reservoir placement, were the cerebellum followed by the occipital lobe. Thereafter enhancement of CN VII/VIII and the ependymal lining of the lateral ventricles were more frequently noted. The sensitivity of MRI for the detection of LMD varies in the literature between from 53% and 79% [2,18–20]. In our study of patients with CSF positive cytology, LMD was detected on MRI in 185 of 270 patients for a sensitivity of 68.5%. CSF collection for cytology was either by lumbar puncture (n = 196) or from an Ommaya reservoir (n = 74). Since our study included only patients with positive CSF cytology, we cannot comment if one method is preferred for diagnosis and could be investigated in the future.

Please cite this article in press as: Debnam JM et al. Most common sites on MRI of intracranial neoplastic leptomeningeal disease. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.07.020

3

J.M. Debnam et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx Table 1 Brain: Frequencies of sites of FLAIR signal hyperintensity. Location

Cerebellum No Yes Occipital No Yes Parietal No Yes Temporal No Yes Frontal No Yes

No. of patients (n = 270)

Percentage

P-value

174 96

64.44 35.56

0.56

178 92

65.93 34.07

0.003

195 75

72.22 27.78

0.006

211 59

78.15 21.85

0.02

225 45

83.33 16.67

Overall conclusion: Cerebellum > Occipital  Parietal  Temporal  Frontal. ‘‘>” means higher but not statistically significant. ‘‘  ” means significantly higher.

Table 2 Brain: Frequencies of sites of leptomeningeal enhancement.

Location

Fig. 1. Common locations of neoplastic LMD in the intracranial compartment. A. Sagittal T1 post-contrast MRI demonstrates sulcal enhancement in the cerebellum and superior vermis (arrow). B. Axial T1 post-contrast MRI with enhancement of CN VII/VIII (arrows). C. Axial T1 post-contrast MRI illustrating leptomeningeal enhancement in the dependent portions of the brain, including the interpeduncular cistern (large arrow), superior vermis (long arrow), and dependent posterior temporal and occipital lobes (small arrow). D. Coronal T1 post-contrast MRI showing enhancement along the ependymal lining of the lateral ventricles (arrows). E. Axial FLAIR sequence with signal hyperintensity involving the superior cerebellar vermis and the supratentorial compartment (arrows).

Parenchymal brain metastases have been reported to be associated with LMD in 21–82% of cases of parenchymal metastasis [18– 21]. Our results were similar and identified 108 (40%) of the patients with LMD had parenchymal metastasis. Of these 108 patients, 83 (76.9%) had known brain metastasis prior to the development of LMD while in 25 (23.1%) patients the detection of new brain metastasis occurred concurrently with the diagnosis of LMD. This coincides with other reports in the literature that LMD occurs later in the course of patient’s disease [10,11]. In the literature, the cancers most commonly associated with LMD are lymphoma, leukemia, breast cancer, lung cancer, and melanoma [9,22,23]. In other studies, solid tumors most commonly giving rise to LMD are breast cancer (12–35%), lung cancer (10– 26%), and melanoma (5–25%) [1–3,6,24]. In our study, the cancers most commonly associated with LMD are similar to those in the literature. Lymphoma (31%) was the most frequent cancer among those demonstrating LMD, followed by breast cancer (24%), lung cancer (11%), and melanoma (11%).

Cerebellum No Yes Occipital No Yes Parietal No Yes Temporal No Yes Frontal No Yes

No. of patients (n = 270)

Percentage

P-value

188 82

69.63 30.37

0.051

203 67

75.19 24.81

0.009

219 51

81.11 18.89

0.45

223 47

82.59 17.41

0.73

225 45

83.33 16.67

Overall conclusion: cerebellum > occipital>>parietal > temporal > frontal. ‘‘ > ” means higher but not statistically significant. ‘‘” means significantly higher.

The median survival of untreated LMD is 4–6 weeks, as LMD portends a poor prognosis even with aggressive treatment. Even with combined treatment, survival is usually less than 8 months, with a median survival of 2–3 months [25]. The patients in our study had a similar median survival of 4 months. Tumor cells can gain access to the CSF in several ways. The most common routes are likely hematogenous, through arachnoid vessels, or direct extension from the brain. Leptomeningeal spread of disease also disseminate from choroid plexus metastases spreading into the CSF or from dural-based metastasis [25]. On imaging, leptomeningeal enhancement has been reported in dependent regions [26], but specific locations where LMD is likely to occur have not been described. In our study, findings of FLAIR signal hyperintensity and enhancement predominantly involved the sulci of the posterior fossa and dorsal supratentorial compartment in gravity-dependent regions, namely the occipital lobe and may support the theory of a postural and gravitation-dependent tumor cell deposition and collection. Another possible explanation may be related to the relationship of the occipital lobe and cerebellum with respect to the flow of CSF to and from the spinal canal. Thus it likely that the posterior fossa and occipital lobes are in con-

Please cite this article in press as: Debnam JM et al. Most common sites on MRI of intracranial neoplastic leptomeningeal disease. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.07.020

4

J.M. Debnam et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx

Table 3 Other sites: Frequencies of leptomeningeal enhancement.

Location CNVII/VIII No Yes Ependymal No Yes Basal cisterns No Yes Meckel’s cave No Yes CN V No Yes Pituitary No Yes CN III No Yes

No. of patients (n = 270)

Percentage

P-value

215 55

79.63 20.37

0.28

224 46

82.96 17.04

0.33

231 39

85.56 14.43

0.37

237 33

87.78 12.22

0.24

243 27

90 10

0.02

257 13

95.19 4.81

0.37

261 9

96.67 3.33

Overall Conclusions: CNVII/VIII > ependymal > basal cistern > Meckel’s cave > CNV  Pituitary > CNIII. CNVII/VIII  basal cisterns (p = 0.02). ependymal and basal cisterns  5CN (p = 0.02, 0.05, respectively). ‘‘>” means higher but not statistically significant. ‘‘” means significantly higher.

stant contact with CSF relative to the other lobes. In addition, retrograde CSF flow from outside the ventricular system could deposit tumor cells along the ependymal surface and explain the relatively common ependymal disease involvement. Another route of leptomeningeal spread of disease is from the CNs into the subarachnoid space [25]. In our study, CN VII/VIII complex was involved more often than CN V and CN III. CN V is the largest of the CNs and one would expect it would be more susceptible due to larger surface area to leptomeningeal seeding. However, CN VII/VIII complex resides in a bony canal and Enzmann et al. [26] report that CSF within the internal auditory canal pulsates less than CSF within the adjacent cerebellopontine angle [26], We therefore postulate that a lack of CSF flow and/or eddying of CSF creates backwaters in this location facilitating collection, adherence and growth of tumor cells on the CNs. LMD presenting exclusively as CN involvement occurred in only 10 (3.7%) patients in our study and highlights the need to be vigilant and thoroughly assess the cranial nerves when there is clinical suspicion for LMD. Tumor cells may find sanctuary in the meninges and CSF compartment because many systemic chemotherapeutic agents fail to cross the intact blood-CSF barrier, and tumor cells may proliferate before the instillation of intrathecal chemotherapy [25]. Intrathecal chemotherapy targets the CSF space and may be administered through repeat lumbar punctures or an Ommaya reservoir. Administration of intrathecal chemotherapy via an Ommaya reservoir is preferred, as it ensures a more homogeneous distribution of chemotherapy over lumbar puncture [7,27]. Future studies could assess MRI before and after initiation of intrathecal chemotherapy via an Ommaya reservoir to determine the efficacy of this treatment relative to systemic chemotherapy. Most of the patients had some form of treatment before the MRI. As LMD often occurs as a late complication of the disease course [10,11] we do not feel that this is a true limitation but rather takes into account the typical presentation of LMD. Thickened enhancement of the meninges may occur after craniotomy [28,29], and LMD has been reported following the resection of cerebellar metastases, possibly owing to malignant cells spilling

into the CSF [30,31]. Similarly, radiation [32] and chemotherapy [33,34] can affect the leptomeningeal enhancement pattern, simulating LMD or progressive disease. MRI is ideally acquired before lumbar puncture as the procedure may irritate the meninges, leading to pachymeningeal, but not leptomeningeal enhancement [35]. Therefore even though diagnostic lumbar puncture was performed before MRI in 196 of the 270 patients (73%), we do not feel that the possibility of false positives confound our results. Additionally all of the patients in our study with positive CSF cytologic findings had a lumbar puncture within 30 days of the MRI. In our study, the patients had an MRI at varying field strengths, either 1.5 T or 3 T. The 3 T MRI, used in 47 patients (17%), has higher contrast-to-noise ratio, which can increase the signal intensity. All patients underwent routine MRI of the brain performed at a 5-mm thickness. Conceivably, we may be underestimating the amount of LMD affecting the CNs, especially CN III. Future studies could include patients with de novo, untreated LMD to assess whether our findings for the most common sites of LMD are similar to in this population. Also, 1.5 T MRI could be compared with 3 T MRI to determine whether 3 T MRI is more sensitive or whether an increased contrast-to-noise ratio produces more false-positive findings. 5. Conclusion Neoplastic LMD affects the brain and CNs diffusely. On imaging, in patients who are to be treated for LMD with an Ommaya, LMD occurs more often in the supine posture gravitationally dependent portions of the brain parenchyma, predominantly the cerebellar folia and occipital lobes, rather than in the remainder of the supratentorial compartment. CN VII/VIII complex and the ependymal lining of the lateral ventricles are more frequently involved with LMD than the other CNs, the basal cisterns or the pituitary stalk; and the CNs may also be the only site of LMD on imaging. Patients with LMD appear to commonly have parenchymal metastasis. Further study will be necessary to determine the pathophysiology and significance of this distribution pattern of LMD. Acknowledgment We thank Sarah Bronson for editorial assistance with the manuscript. Funding None. Conflict of interest statement The authors have no conflicts of interest to report. References [1] Kaplan JG, DeSouza TG, Farkash A, et al. Leptomeningeal metastases: comparison of clinical features and laboratory data of solid tumors, lymphomas and leukemias. J Neurooncol 1990;9(3):225–9. [2] Clarke JL, Perez HR, Jacks LM, et al. Leptomeningeal metastases in the MRI era. Neurology 2010;74(18):1449–54. [3] Kesari S, Batchelor TT. Leptomeningeal metastases. Neurol Clin 2003;21 (1):25–66. [4] Little JR, Dale AJ, Okazaki H. Meningeal carcinomatosis. Clinical manifestations. Arch Neurol 1974;30(2):138–43. [5] Olson ME, Chernik NL, Posner JB. Infiltration of the leptomeninges by systemic cancer. A clinical and pathologic study. Arch Neurol 1974;30(2):122–37. [6] Posner JB, Chernik NL. Intracranial metastases from systemic cancer. Adv Neurol 1978;19:579–92.

Please cite this article in press as: Debnam JM et al. Most common sites on MRI of intracranial neoplastic leptomeningeal disease. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.07.020

J.M. Debnam et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx [7] Shapiro WR, Young DF, Mehta BM. Methotrexate: distribution in cerebrospinal fluid after intravenous, ventricular and lumbar injections. N Engl J Med 1975;293(4):161–6. [8] Theodore WH, Gendelman S. Meningeal carcinomatosis. Arch Neurol 1981;38 (11):696–9. [9] Wasserstrom WR, Glass JP, Posner JB. Diagnosis and treatment of leptomeningeal metastases from solid tumors: experience with 90 patients. Cancer 1982;49(4):759–72. [10] Omar AI, Mason WP. Leptomeningeal metastases. In: Neurologic Complications of Cancer. New York: Oxford University Press, Inc; 2009. p. 240–81. [11] Drappatz J, Batchelor T. Leptomeningeal metastasis. In: American Society of Clinical Oncology educational book. Pp. 100–105. [12] Brem SS, Bierman PJ, Brem H, et al. National Comprehensive Cancer Network, Central nervous system. J Natl Compr Canc Netw 2011;9(4):352–400. [13] Glantz MJ, Cole BF, Glantz LK, et al. Cerebrospinal fluid cytology in patients with cancer: minimizing false-negative results. Cancer 1998;82(4):733–9. [14] Lossos A, Siegal T. Spinal subarachnoid hemorrhage associated with leptomeningeal metastases. J Neurooncol 1992;12(2):167–71. [15] Klein P, Haley EC, Wooten GF, et al. Focal cerebral infarctions associated with perivascular tumor infiltrates in carcinomatous leptomeningeal metastases. J Neurooncol 1992;12(2):167–71. [16] DeAngelis LM. Current diagnosis and treatment of leptomeningeal metastasis. J Neurooncol 1998;38(2–3):245–52. [17] Lin N, Dunn IF, Glantz M, et al. Benefit of ventriculoperitoneal cerebrospinal fluid shunting and intrathecal chemotherapy in neoplastic meningitis: a retrospective, case-controlled study. J Neurosurg 2011;115(4):730–6. [18] Park JH, Kim YJ, Lee JO, et al. Clinical outcomes of leptomeningeal metastasis in patients with non-small cell lung cancer in the modern chemotherapy era. Lung Cancer 2012;76(3):387–92. [19] Rudnicka H, Niwin´ska A, Murawska M. Breast cancer leptomeningeal metastasis: The role of multimodality treatment. J Neurooncol 2007;84 (1):57–62. [20] Morris PG, Reiner AS, Szenberg OR. Leptomeningeal metastasis from non-small cell lung cancer: survival and the impact of whole brain radiotherapy. J Thorac Oncol 2012;7(2):382–5.

5

[21] Clatot F, Philippin-Lauridant G, Ouvrier MJ. Clinical improvement and survival in breast cancer leptomeningeal metastasis correlate with the cytologic response to intrathecal chemotherapy. J Neurooncol 2009;95(3):421–6. [22] Gleissner B, Chamberlain MC. Neoplastic meningitis. Lancet Neurol 2006;5 (5):443–52. [23] Groves M. Leptomeningeal metastases: Still a challenge. In: American Society of Clinical Oncology educational book. 2008:80–87. [24] Byrne TN, Cascino TL, Posner JB. Brain metastasis from melanoma. J Neurooncol 1983;1(4):313–7. [25] Le Rhun E, Taillibert S, Chamberlain MC. Carcinomatous meningitis: leptomeningeal metastases in solid tumors. Surg Neurol Int 2013;4(Suppl. 4):S265–88. [26] Enzmann DR, Rubin JB, DeLaPaz R, et al. Cerebrospinal fluid pulsation: benefits and pitfalls in MR imaging. Radiology 1986;161(3):773–8. [27] Dietemann JL, Correia Bernardo R, Bogorin A, et al. Normal and abnormal meningeal enhancement: MRI features. J Radiol 2005;86(11):1659–83. [28] Burke JW, Podrasky AE, Bradley Jr WG. Meninges: benign postoperative enhancement on MR images. Radiology 1990;174(1):99–102. [29] Leach JL, Jones BV, Tomsick TA, et al. Normal appearance of arachnoid granulations on contrast-enhanced CT and MR of the brain: differentiation from dural sinus disease. AJNR Am J Neuroradiol 1996;17(8):1523–32. [30] Norris LK, Grossman SA, Olivi A. Neoplastic meningitis following surgical resection of isolated cerebellar metastasis: a potentially preventable complication. J Neurooncol 1997;32(3):215–23. [31] DeAngelis LM, Mandell LR, Thaler HT, et al. The role of postoperative radiotherapy after resection of single brain metastases. Neurosurgery 1989;24(6):798–805. [32] Muscal JA, Jones JY, Paulino AC, et al. Changes mimicking new leptomeningeal disease after intensity-modulated radiotherapy for medulloblastoma. Int J Radiat Oncol Biol Phys 2009;73(1):214–21. [33] Weiss HD, Walker MD, Wiernik PH. Neurotoxicity of commonly used antineoplastic drugs. N Engl J Med 1974;291(2):75–81. [34] Jacob LA, Sreevatsa A, Chinnagiriyappa LK, et al. Methotrexate-induced chemical meningitis in patients with acute lymphoblastic leukemia/lymphoma. Ann Indian Acad Neurol 2015;18(2):206–9. [35] Mittl Jr RL, Yousem DM. Frequency of unexplained meningeal enhancement in the brain after lumbar puncture. AJNR Am J Neuroradiol 1994;15(4):633–8.

Please cite this article in press as: Debnam JM et al. Most common sites on MRI of intracranial neoplastic leptomeningeal disease. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.07.020