Volume=20 36, Supplement=20 2, Pages S55-S68 (August = 2009)

9 of 10

ABSTRACT

FULL=20 TEXT

FULL-TEXT PDF (658=20 KB)

The Role of Intra-cerebrospinal = Fluid=20 Treatment and Prophylaxis in Patients With Solid Tumors

Riccardo Soffiettia 3D"Corresponding

, Wallace Akerleyb, = Randy=20 L. Jensenc, = Joachim Bischoffd, = Anne=20 Constanze Regierere

Metastasis to the central = nervous=20 system (CNS), including neoplastic meningitis (NM), is a = devastating=20 complication of systemic cancer. With the improved survival = of=20 cancer patients, the incidence of CNS metastasis is rising,=20 especially among those with breast or lung carcinoma. New = therapies=20 that effectively treat these primary tumors outside of the = CNS have=20 underscored the significance of CNS metastases; they have = become a=20 significant clinical issue and a therapeutic challenge. This = review=20 discusses clinical situations in which treatment or = chemoprophylaxis=20 of CNS metastases and NM from breast or lung cancer may play = an=20 important role. Potential clinical trials to assess these=20 assumptions also will be proposed.

Article Outline

=E2=80=A2 Abstract

=E2=80=A2 NM=20 as a Complication of Breast Cancer

=E2=80=A2 Intra-CSF=20 Chemotherapy: Improving Clinical Outcome for Breast Cancer = Patients=20 With NM

=E2=80=A2 The=20 Usefulness of Prophylactic Intra-CSF Treatment in Patients = With=20 Breast Cancer and CNS Metastases

=E2=80=A2 Identifying=20 Patients With Breast Cancer at High Risk of Developing CNS=20 Metastases

=E2=80=A2 Is=20 There a Subpopulation of Breast Cancer Patients for Whom CNS = Prophylaxis May Be Justified?

=E2=80=A2 NM=20 as a Complication of Lung Cancer

=E2=80=A2 CNS=20 Metastases in Patients With Non-Small Cell Lung = Cancer

=E2=80=A2 Identifying=20 High-Risk Patients Who May Be Candidates for CNS = Prophylactic=20 Treatment

=E2=80=A2 CNS=20 Metastases in Patients With SCLC

=E2=80=A2 Postsurgical=20 Prophylaxis

=E2=80=A2 Conclusions

=E2=80=A2 References

=E2=80=A2 Copyright

It is estimated that in 2008, there = were more=20 than 1.4 million new cases of cancer in the United States, = with=20 565,650 deaths.1 = Between 1973 and=20 2003, Barnholtz-Sloan and colleagues reported a 9.6% = incidence of=20 central nervous system (CNS) metastases among 16,210 = patients=20 diagnosed with solid tumors in the Metropolitan Detroit = Cancer=20 Surveillance System.2 Lung = cancer was=20 reported to be the solid tumor that relapsed to the CNS most = frequently (19.9%), followed by melanoma (6.9%), renal = (6.5%),=20 breast (5.1%), and colorectal cancers (1.8%).

Neoplastic meningitis (NM) is a = devastating=20 form of CNS metastasis characterized by the invasion of the=20 leptomeninges and cerebrospinal fluid (CSF) by malignant = cells.3 NM = is associated=20 with rapid and progressive neurological deterioration, and = median=20 survival in untreated patients with NM is only a few = weeks.4 The = actual=20 overall incidence of NM is difficult to establish, but it = has been=20 estimated to be between 6% and 21% in patients with solid = tumors.5 The = main solid=20 tumors to invade the leptomeninges are breast and lung = cancer, which=20 account for 12% to 50% and 10% to 26%, respectively, of = cases of=20 NM.6, 7, 8 Many = cases go=20 unrecognized. The chance of developing NM also may increase = in=20 certain situations, such as surgical resection of brain = metastases=20 located in the posterior fossa.9, 10, = 11, = 12

As detailed elsewhere in this = supplement, the=20 diagnosis of NM relies on either positive CSF cytology, = abnormal=20 magnetic resonance imaging (MRI), or clinical findings in = the=20 appropriate circumstances (eg, raised protein and cranial = nerve=20 palsy). Since the presentation of NM is protean, it is = important to=20 have a high clinical suspicion in the presence of vague = neurological=20 symptoms. Failure to evaluate is common because of = unrecognized=20 symptoms and a perception that there are no available = therapies.

Overall, the occurrence of CNS = metastases,=20 including NM, has been increasing.5 This = is due to a=20 number of factors. First, the development of improved = treatments for=20 primary breast and lung cancers provides more effective = control of=20 systemic disease and results in longer survival.13 = However,=20 because of the relative inability of systemic chemotherapy = to=20 penetrate the blood-brain barrier (BBB), a sanctuary site = for=20 metastatic cells is created in the CNS.5 For = example, a=20 retrospective study by Clayton and colleagues demonstrated = that 25%=20 of patients with breast cancer (n =3D 93) treated with = trastuzumab=20 developed brain metastases, and suggested that although = trastuzumab=20 promoted long-term survival, it did not penetrate the CSF in = sufficient quantities, creating a metastatic sanctuary = site.14 A = 2-year study=20 of patients with small cell lung cancer (SCLC) in the = Netherlands=20 reported a 2% incidence of NM at the time of primary cancer=20 diagnosis.15 = Despite=20 combination chemotherapy consisting of cyclophosphamide,=20 doxorubicin, and etoposide, the 2-year cumulative incidence = of NM on=20 follow-up was 10%,15 = suggesting=20 that the CSF may have acted as a sanctuary site for = metastasized=20 cells. Further, nine of 21 (43%) patients with metastatic=20 adenocarcinoma of the lung responding to an epidermal growth = factor=20 receptor (EGFR) inhibitor developed brain metastases, even = though=20 the majority demonstrated control of extracranial disease.16 = Second, an=20 increased awareness of this complication and improved = diagnostic=20 modalities have resulted in increased detection of CNS = metastases.5 For = some=20 patients, CNS involvement may be the only cause of decreased = quality=20 of life (QoL) and the main threat to survival.

Although the treatment of NM is = palliative,=20 treatment may improve time to neurological progression and = QoL.=20 Conventional treatment for patients with brain metastases is = whole-brain radiation therapy (WBRT).17 = Although WBRT=20 may relieve symptoms, survival averages only a few months.17 = For patients=20 with limited or no systemic disease and good neurological = status,=20 the combination of surgery or radiosurgery and WBRT results = in=20 improved outcome (eg, improved local control and delayed=20 intracranial recurrence) compared with either treatment = alone.17, = 18, = 19

Systemic chemotherapies (eg, = methotrexate=20 [MTX]) given at high doses to overcome the BBB have shown = some=20 activity in treating NM but may lead to substantial systemic = toxicity.20, = 21 = Intra-CSF=20 chemotherapies circumvent the BBB and treat the entire = neuroaxis=20 while minimizing systemic toxicity. Agents that are most = commonly=20 used in clinical practice for intrathecal administration are = MTX,=20 cytarabine, thiotepa, and sustained-release cytarabine = (liposomal=20 cytarabine).22 = MTX and=20 cytarabine are indicated for the treatment of lymphomatous = and=20 leukemic meningitis, and liposomal cytarabine is indicated = for the=20 treatment of lymphomatous meningitis. However, none of these = agents=20 is approved by the US Food and Drug Administration for the = treatment=20 of NM from solid tumors.

Few clinical trials assessing the = efficacy of=20 intra-CSF therapy for patients with solid-tumor = leptomeningeal=20 metastases have been conducted, and only three are = prospective,=20 randomized trials (Table 1).8, 23, = 24, = 25, = 26, = 27, = 28, = 29, = 30 = However, a=20 retrospective review of medical records from 126 patients = with=20 cytologically confirmed NM from solid tumors demonstrated = that=20 intra-CSF chemotherapy was a significant predictive factor = of=20 survival (P =3D .01).4 Some = studies=20 also indicate that NM from breast or lung cancer may respond = to=20 intra-CSF therapy,25, = 29 = and long-term=20 survival has been observed occasionally in patients with = breast=20 cancer and leptomeningeal metastases (ie, >112 weeks).8, 23, = 25 = Still, the=20 benefits of chemotherapy in NM from solid tumors are not yet = clear.30, = 31, = 32

Table 1.

Efficacy of Intrathecal = Therapies=20 for the Treatment of Solid=20 Tumors


First Author, Year	Study Description	Treatments	Response	Median Survival	Primary Tumor Data
Yap, 198223	N=20 =3D 40 Breast, n =3D=20 40	WBRT + MTX	65% Negative CSF cytology + = normalization of=20 CSF biochemistry + improvement of neurological=20 status	Responders: 6 = mo Non-responders: 1=20 mo
Wasserstrom, 19828	N=20 =3D 90 Breast, n =3D = 46 Lung, n =3D=20 23	MTX	50% Stabilization/improvement of = neurological=20 symptoms =E2=89=A51 = mo	5.8 mo (1-29)	Survival: Breast: median 7.2 mo = (1-29) Lung: median 4 mo = (1-10) Other: = 1-12
Ongerboer de Visser, = 198324	N=20 =3D 33 NR Breast, n = =3D33 Prospective,=20 NR	MTX/Ommaya RT/MTX/lumbar Untreated		6=20 mo (0.5-28) 1=20 mo (0.25-7) 2=20 mo (0.25-3)
Hitchins, 198725	N=20 =3D 44 R Breast (n =3D = 11) SCLC (n =3D=20 13)	MTX MTX + = AraC	61% 45% (P =3D = NS) Improved clinical status + = negative CSF=20 cytology + normalization of CSF biochemistry for = =E2=89=A54=20 wk	Responders: 18 = wk Non- responders: 7=20 wk (P=20 <.05)	Response = rate: Breast: 36% SCLC: 69% ACUP: 50% Survival: Breast: median 9 wk (59 for responders)=20 (1-112) SCLC: median 8 wk = (1-26) ACUP: median 8 wk=20 (1-39)
Grossman, 199326	R N=20 =3D 52 Breast, n =3D 25 Lung, n =3D=20 12	MTX Thiotepa	0% Normal neurologic examination + = negative=20 CSF cytology (lumbar and ventricular) + no meningeal = masses +=20 normal CSF = biochemistry	15.9 wk 14.1 = wk	Survival: Breast: median 15 wk (one patient >110=20 wk) Lung: median 8 wk Other solid tumors: median 17=20 wk
Glantz, 199927	N=20 =3D 61 R Breast, n =3D = 22 Lung, n = =3D10 Other, n =3D=20 29	Liposomal = AraC MTX	26% 20% (P =3D = NS) Negative cytology at all = previously >0=20 sites (lumbar and ventricular) + stable or improved=20 neurological = condition	105 days 78=20 days (P =3D=20 NS)	Response: Breast: liposomal AraC: 2/11, MTX:=20 0/11 SCLC: liposomal AraC: 0/2, MTX:=20 2/2 NSCLC: liposomal AraC: 0/3, MTX:=20 1/3 Other: liposomal AraC: 6/15, MTX:=20 3/14 One patient with breast cancer treated with = liposomal=20 AraC survived 465 = days
Jaeckle, 200228	N=20 =3D 110 Breast, n =3D = 38 Lung, n =3D=20 18	Liposomal AraC	27% Negative cytology at all = previously >0=20 sites (lumbar and ventricular) + stable or improved=20 neurological = condition	95 days (7-791+)
Kim, 200329	N=20 =3D 50 NR Breast (n =3D = 13) Lung (n =3D = 33) Other	MTX MTX+ hydrocortisone +=20 AraC	13.8% 38.5% (P = =3D.036) Negative cytology from lumbar=20 CSF	10.4 wk 18.6 wk P=20 =3D.029	Response: Breast: 36% SCLC: = 80%
Boogerd, 200430	N=20 =3D 35 Breast, n =3D=20 35	MTX + RT (if=20 necessary) RT=20 but no IT treatment	59% 67% Stable or improved neurological=20 condition	18.3 wk 30.3 wk (P =3D=20 NS)

Abbreviations: ACUP,=20 adenocarcinoma unknown primary; AraC, cytarabine; CSF, = cerebrospinal fluid; IT, intrathecal; MTX, = methotrexate; NR,=20 non-randomized; NSCLC, non-small cell lung cancer; NS, = not=20 significant; R, randomized; RT, radiation therapy; = SCLC, small=20 cell lung cancer; WBRT, whole brain radiation=20 therapy.

Study included = individuals with=20 lymphomatous meningitis. More patients randomized to = MTX (61%)=20 v thiotepa (33%) had breast cancer and were = without=20 evidence of systemic cancer. Radiation was = administered to=20 mass lesions or symptomatic=20 sites.

The increased frequency of CNS = metastases,=20 including NM, is a significant clinical issue and a = therapeutic=20 challenge both with regard to treatment and prevention. CNS=20 prophylaxis in subpopulations at high risk of CNS = metastases, and/or=20 better screening and early diagnosis and treatment, may = improve=20 clinical outcome, including QoL. This underscores the need = to=20 identify patients most at risk of developing CNS = metastases.

The purpose of this review is both = practical=20 and forward-looking. We will first examine clinical = situations in=20 which treatment or chemoprophylaxis of CNS metastases from = breast=20 carcinoma or lung cancer may play an important role. Next, = we will=20 present important considerations for preventing CNS = metastases in=20 the surgical resection setting. We will conclude with a = potential=20 trial proposal and discuss the assumptions presented.

NM as a Complication = of Breast=20 Cancer

In 2008, more than 182,000 women in = the United=20 States were expected to be diagnosed with breast cancer.1 = Historically, 5%=20 to 16% of women with breast cancer develop CNS metastases,2, 33, = 34 = although twice=20 as many cases (30%) have been reported in autopsy series.35 = Among solid=20 tumors, breast cancer is the most common tumor to invade the = leptomeninges.7, 36 = The incidence=20 of NM in patients with breast cancer is approximately 3% to = 8%.33 = However, the=20 incidence has been reported to be 19% upon autopsy = analysis.35

While the advent of improved primary = chemotherapies has led to increased survival and a = consequent=20 increase in the incidence of CNS metastases, patients with = otherwise=20 stable extracranial disease may warrant monitoring and = prophylaxis=20 of CNS relapse. The role of intra-CSF chemotherapy for CNS = disease=20 in breast cancer is unclear. Critical issues to consider = include:=20 (1) the need to determine the capacity of intra-CSF = chemotherapy to=20 improve the clinical outcome of patients with breast cancer = and=20 clinically diagnosed NM with or without CNS metastases (does = it=20 work?); (2) whether patients with breast cancer and CNS = metastases=20 should receive chemoprophylaxis to curtail the risk of = developing=20 subsequent NM (can it be prevented?); (3) the identification = of a=20 subgroup of patients who may benefit from intensive NM = screening but=20 for whom intra-CSF chemoprophylaxis may not be justified = (are there=20 subgroups who are at special risk?); and (4) the use of risk = factors=20 in patients with breast cancer (defined by clinical profile = or tumor=20 biology or systemic treatment) to identify patients for whom = intra-CSF prophylaxis may be beneficial enough to be = investigated in=20 a clinical trial.

Intra-CSF = Chemotherapy:=20 Improving Clinical Outcome for Breast Cancer Patients With=20 NM

Intra-CSF administration of 20 mg = MTX combined=20 with WBRT (30 Gy) resulted in a 65% response rate (>50% = reduction=20 in tumor cell count and/or complete normalization of the = CSF) among=20 40 patients with breast cancer and NM.23 = The median=20 survival time for patients who responded to therapy was 6 = months,=20 compared with 1 month for non-responders (P =3D = .001).23 = Thirty-one=20 percent of patients died from progressive systemic disease = while=20 their NM was in remission. Four patients remained alive and = in=20 remission at 26 to 70+ weeks.23 = Combination=20 therapy with intra-CSF MTX and cytarabine also has been = assessed. In=20 an early randomized study of patients with NM from mostly = solid=20 tumors (N =3D 44), including 11 patients with NM from breast = cancer,=20 intra-CSF MTX resulted in a larger, albeit statistically=20 insignificant, response rate (defined as improved clinical = status,=20 negative CSF cytology, and normal CSF biochemistry for 4 = weeks or=20 longer) than MTX and cytarabine (61% v 45%, = respectively)=20 (P =3D .10).25 = However, the=20 overall median survival was 7 weeks in non-responders and 18 = weeks=20 in responders (P <.05).25 = Among patients=20 with NM from breast cancer, MTX alone resulted in a 67% = response=20 rate, versus a 25% response rate in similar patients treated = with a=20 combination of MTX and cytarabine.25 A = more recent=20 study investigated the combination of intra-CSF MTX, = cytarabine, and=20 hydrocortisone compared with intra-CSF MTX alone in a = non-randomized=20 study that included 13 patients with NM from breast = cancer.29 = The overall=20 response rate among patients with breast cancer was 39%.29 = Treatment with=20 MTX, cytarabine, and hydrocortisone resulted in a markedly, = but not=20 statistically significant, improved survival rate compared = with=20 intra-CSF MTX alone (23.7 weeks v 10.1 weeks, = respectively;=20 P =3D .445). In a study by Wasserstrom and = colleagues, patients=20 with NM from breast cancer were treated with radiation = therapy=20 followed by either intra-CSF MTX or intra-CSF cytarabine. = Patients=20 had a median survival of 7.2 months (range, 1=E2=80=9329 = months), and two=20 patients survived 29 months.8

Intra-CSF MTX and thiotepa were = compared in a=20 randomized study involving 52 patients, of whom 48% had = breast=20 cancer.26 = Most patients=20 deteriorated neurologically within 8 weeks of initiating = therapy.=20 Median survival was 15.9 weeks in the MTX group and 14.1 = weeks in=20 the thiotepa group.26

Liposomal cytarabine, an = extended-release form=20 of cytarabine, was compared with MTX in 61 patients with NM = from=20 solid tumors.27 Of = the 22=20 patients with breast cancer, two of 11 treated with = liposomal=20 cytarabine responded, including one who survived for 465 = days,=20 whereas none of the 11 patients treated with MTX responded = to=20 therapy.27 = Response was=20 defined as a conversion from positive to negative CSF = cytology from=20 lumbar and ventricular sites plus a stable or improved = neurological=20 examination. Using similar response criteria, 28% of 43 = evaluable=20 patients with NM from breast cancer responded to intra-CSF = therapy=20 with liposomal cytarabine.37 = Median time to=20 neurologic progression was 49 days (range, 1=E2=80=93515+ = days), and median=20 survival was 88 days (range, 1=E2=80=93515+ days).37 = Liposomal=20 cytarabine was also investigated in an open-label study of = 110=20 patients with NM, 38 of whom had breast cancer.28 = With similar=20 response criteria, the overall response rate was 27%, with a = median=20 overall survival of 95 days (range, 7=E2=80=93791+ days) and = a median time=20 to neurological progression of 55 days (range, = 0=E2=80=93584+ days).

In addition to intra-CSF therapies, = several=20 case studies have reported favorable response of CNS = metastases38, = 39, = 40 = and NM40, = 41, = 42 to = treatment=20 with oral capecitabine.43, = 44 = Capecitabine=20 is indicated for the treatment of metastatic breast cancer = as=20 monotherapy or in combination with docetaxel.43, = 44 = Based on these=20 encouraging results, a pilot phase II, single-arm, = open-label=20 clinical trial is proposed to test whether combining = capecitabine=20 with liposomal cytarabine will improve the response rate,=20 neurological status, and survival of patients with breast = cancer and=20 NM (Figure 1). This trial could enroll = patients with=20 breast cancer and newly diagnosed CNS involvement (NM with = or=20 without brain metastases). Concurrent treatment such as with = trastuzumab (to treat systemic disease), field radiotherapy = (to=20 treat bulky spinal disease including CSF blocks), and = stereotactic=20 surgery to treat isolated CNS metastases would be allowed. = End=20 points would include neurological progression-free survival = (primary=20 end point), overall response rate based upon clinical = response=20 (improvement of neurological signs and symptoms), cytologic = and=20 parenchymal response, overall survival, and safety. The=20 predisposition for intrathecal medication-related = toxicities, as=20 discussed previously in this supplement, would need careful=20 evaluation.

Figure=20 1. Proposal for a pilot phase II = single-arm,=20 open-label clinical trial of capecitabine and = liposomal=20 cytarabine for the treatment of newly diagnosed = neoplastic=20 meningitis with or without central nervous system = metastases=20 in patients with breast cancer.

The Usefulness of = Prophylactic=20 Intra-CSF Treatment in Patients With Breast Cancer and CNS=20 Metastases

At least half of the patients with = CNS=20 involvement have stable systemic disease, and the brain is = often the=20 first site of recurrence.13 = When a brain=20 relapse is diagnosed, seeding in the leptomeninges or CSF = may=20 already exist, even if it is not yet detectable. In = addition, some=20 patients with CNS metastases may be at increased risk of NM. = Potential risk factors for meningeal relapse include surgery = for=20 isolated CNS metastases, especially surgery of metastases = located in=20 the posterior fossa9, 10, = 11, = 12; = the proximity=20 of brain metastases to the subarachnoid space6; the = number of=20 metastases; the coexistence of vertebral metastases6; the = tumor=20 type36; = and the=20 length of follow-up. Two potential clinical trials that = could test=20 the usefulness of CNS-directed chemoprophylaxis for patients = with=20 breast cancer and brain metastases are outlined in Figure 2. Both trials could include = patients with=20 recursive partitioning analysis class I or II (class I: = patients=20 with a Karnofsky Performance Score [KPS] =E2=89=A570, <65 = years of age,=20 controlled primary tumor and no extracranial metastases; = class III:=20 KPS <70; class II: all others)45 = with complete=20 or partial response after surgery, stereotactic surgery, = radiation=20 therapy, chemotherapy, or combined treatments. The first = trial would=20 randomize patients to observation or chemoprophylaxis with = liposomal=20 cytarabine (Figure 2A). The second trial would = assign=20 patients with positive CSF cytology to treatment with = liposomal=20 cytarabine, while those with negative CSF cytology would be = observed=20 (Figure 2B). Results from these trials = could help=20 determine the value of CNS-directed chemoprophylaxis.

Figure=20 2. (A) Proposal for an intra-cerebrospinal = fluid=20 (CSF) prophylaxis trial in patients with brain = metastases from=20 breast cancer. Stratification for: CSF cytology = (positive=20 v negative v uncertain); tumor site = (posterior=20 fossa v cerebral hemispheres); surgical = resection (yes=20 v no); and vertebral metastases (yes v = no). (B)=20 Proposal for a CSF prophylaxis trial in patients with = brain=20 metastases from breast cancer. CR, complete response; = PR,=20 partial response; MRI, magnetic resonance imaging; = RPA,=20 recursive partitioning analysis (used for prognostic=20 classification); S, surgery; SRS, stereotactic = radiosurgery;=20 WBRT, whole-brain radiation therapy.

Identifying Patients = With=20 Breast Cancer at High Risk of Developing CNS=20 Metastases

The profoundly morbid effects of CNS = metastases and the early identification of patients at risk = may=20 justify early systematic screening for CNS disease in = otherwise=20 asymptomatic patients with breast cancer and/or define a=20 subpopulation of patients that may benefit from prophylaxis. = However, there are currently no studies that address whether = early=20 screening for CNS metastases (parenchymal or meningeal) in = patients=20 with breast cancer would improve survival. Despite the = paucity of=20 data, it is possible to identify certain subpopulations of = patients=20 with breast cancer who are at high risk of developing CNS=20 metastases.

Several factors have been identified = that are=20 associated with a higher risk of developing CNS metastases = (Table 2).46, = 47, = 48, = 49 = These include=20 young age,49, = 50, = 51, = 52 = progressive=20 disease stage,48, = 53 = elevated=20 lactate dehydrogenase,54 = nodal=20 status,53 = lung=20 metastases, and tumor characteristics such as negative = estrogen (ER)=20 or progesterone (PgR) receptor status, human epidermal = growth factor=20 receptor 2 (HER2) overexpression, cytokeratin (CK) 19 = expression,=20 tumor size, and grade.14, = 46, = 47, = 48, = 49, = 50, = 52, = 53, = 55, = 56 By = evaluating=20 several factors concomitantly, it may be possible to = identify a=20 group of women with a much higher risk of CNS metastases. = This is=20 illustrated in a study by Evans and colleagues, in which = young age=20 (ie, <50 years old) combined with ER=E2=88=92 status resulted in a = 53% risk of=20 developing CNS metastases.52 = For patients=20 with several risk factors, intensive screening (serial MRI = and CSF=20 cytology) may be justified to detect early signs of CNS = disease.=20 Even so, a true definition of risk factors still remains = unclear,=20 and the optimal timing and frequency of screening has not = been=20 established. Of consideration are the cost and inconvenience = of=20 routine CNS screening and the fact that some procedures are = invasive=20 (eg, lumbar puncture). In addition, the clinical and = biological=20 significance of isolated cells in the CSF is unclear and = needs to be=20 better defined.

Table 2.

Risk Factors for = Central Nervous=20 System Relapse in Patients With Breast Cancer


First Author, Year	HER-2 Overexpression	Negative Receptor Status	Large Tumor Size	CK 5/6/19 Expression	Infiltrating Ductal Carcinoma = Histology	Lung Metastases	Disease Stage
Slimane, 200446		HR =3D 4.2; P = =3D .002				HR =3D 4.3; P = =3D 0.0003
Gabos, 200647	HR =3D 3.55; P = =3D .006	ER+: HR =3D 0.32; P = =3D .002	=E2=89=A52=20 cm HR=20 =3D 2.76; P =3D = .013
Souglakos, 200648	HR =3D 29.4; P=20 <.0001			CK19: HR =3D 8.31; = P =3D=20 .001			HR =3D 0.23; = P=20 <.0001
Tham, = 200649		HR =3D = 2.8;=20 P <.001			HR =3D = 2.5;=20 P =3D.02

Abbreviations: CK, = cytokeratin;=20 HER, human epidermal growth factor receptor; HR, = hazard=20 ratio.

Identified by = multivariate=20 = analysis.

Is There a = Subpopulation of=20 Breast Cancer Patients for Whom CNS Prophylaxis May Be=20 Justified?

To be a candidate for CNS = prophylaxis, the=20 patient with breast cancer should be at high risk of CNS = metastases=20 and have a reasonable life expectancy. Analysis of breast = tumor gene=20 expression patterns may help identify patients who are = candidates=20 for CNS/CSF prophylaxis. More aggressive breast cancers, = such as=20 those with HER2/neu overexpression, and triple-negative = receptor=20 status (ER=E2=88=92, = PgR=E2=88=92, and HER2=E2=88=92), have an increased = risk of CNS=20 metastasis and are associated with a poorer systemic = prognosis.46, = 47, = 48, = 49, = 52, = 53, = 55, = 57

HER2+ breast=20 cancer can be effectively treated with a = trastuzumab-containing=20 regimen, resulting in improved survival.58, = 59 = While=20 trastuzumab is effective at eradicating systemic tumors, it = may not=20 prevent CNS metastases. As a monoclonal antibody directed = against=20 the extracellular domain of the HER2 receptor, the high = molecular=20 weight of trastuzumab may prevent it from crossing the BBB. = As=20 briefly mentioned earlier in this review, this may partly = explain=20 why brain metastases are increasingly reported among = patients with=20 breast cancer who are treated with a trastuzumab-based = regimen (Table 3). Approximately 25% to 48% of = women=20 treated with trastuzumab develop CNS metastases,13, = 14, = 55, = 60, = 61, = 62, = 63, = 64, = 65 = including 5%=20 with NM alone13 = and 14% to 24%=20 with NM together with parenchymal metastases.13, = 62

Table 3.

Frequency of Central = Nervous=20 System Metastases in Relation to Human Epidermal = Growth Factor=20 Receptor 2 Overexpression and Survival in Patients = With Breast=20 Cancer


First Author, Year	No. of Patients	Patients With CNS Metastases n (%)	Patients With HER2 Overexpressing Tumors = (%)	Median Survival (mo)
Bendell, 200313	122	42(34)	=E2=80=94	13
Lower, 200360	87	22(25)	81	=E2=80=94
Van Pelt, 200361	22	4(18)	100	=E2=80=94
Clayton, 200414	93	23(25)	94	5.4
Lai, 200462	79	38(48)=E2=80=A0	47	24.9
Shmueli, 200463	32	10(31)	100	=E2=80=94
Burstein, 200555	234	23(10)=E2=80=A1	100	=E2=80=94
Stemmler, 200664	136	42(31)	100	13
Yau, = 200665	87	23(30)	100	4

Abbreviations: CNS, = central=20 nervous system; HER, human epidermal growth factor=20 receptor.

5% with neoplastic = meningitis (NM)=20 alone and 14% with NM and parenchymal brain = metastases.

=E2=80=A0

24% with NM.

=E2=80=A1

Isolated CNS=20 progression.

The tyrosine kinase inhibitor = lapatinib, in=20 combination with capecitabine, has been shown to be active = against=20 systemic disease and brain metastasis in women with = advanced,=20 progressive HER2+ breast = cancer who=20 have received multiple previous treatments.66, = 67 = Anecdotal=20 evidence indicates a potential role for capecitabine in = NM.40, = 41, = 42 In = addition,=20 preliminary data of an expanded phase II trial suggest that=20 lapatinib alone is active in patients with HER2+ breast cancer and brain = metastases.68 = Based on these=20 premises, one of the first worldwide trials that = incorporates=20 tyrosine kinase inhibitors in the treatment of NM has been = initiated=20 (J. Bischoff, personal written communication, October 2008). = This=20 multicenter, open-label, single-arm, phase II study assesses = the=20 efficacy and safety of oral lapatinib in combination with = oral=20 capecitabine and intrathecally administered liposomal = cytarabine for=20 the treatment of meningeal and brain metastases in patients = with=20 HER2+ breast cancer (J. = Bischoff,=20 personal written communication, October 2008).

Despite the profound morbidity and = the=20 relatively high occurrence of CNS metastases in patients = with breast=20 cancer, the priority for patients with very aggressive and=20 unresponsive breast tumors, such as basal-like breast = cancers,=20 remains finding therapies that will control the primary = systemic=20 cancer. Before a clinical trial assessing the therapeutic = benefit of=20 CNS prophylaxis in patients with breast cancer is performed, = it is=20 necessary to better identify patients at risk and develop = superior=20 regimens and/or therapies to treat CNS metastases from = breast=20 cancer.

NM as a Complication = of Lung=20 Cancer

Lung cancer is the leading cause of = cancer=20 mortality among both men and women in the United States and = is=20 estimated to be responsible for 29% of all cancer deaths in = 2008.1 = Mortality is due=20 to end-organ failure caused by metastases, and CNS = metastases are a=20 common problem in lung cancer. According to the Metropolitan = Detroit=20 Cancer Surveillance System study, the incidence of brain = metastases=20 is 19.9% in patients diagnosed with primary lung cancer2 and = as high as=20 30% in patients with advanced-stage lung cancer.2 As = with breast=20 cancer, effective multimodality regimens integrating = chemotherapy=20 with radiation therapy and/or surgery have increased the = survival of=20 patients with lung cancer. Three-year survival is now = possible for=20 patients with metastatic lung cancer, and 1-year survival = can reach=20 50%.69 As = with breast=20 cancer, the longer patients with lung cancer live, the = higher the=20 chance they will develop CNS metastases. Additionally, the = routine=20 use of positron-emission tomography scanning in individuals = with=20 lung cancer allows for more precise staging of lung cancer, = which=20 should eventually translate into greater survival. MRI = screening has=20 become routine in both early-stage and metastatic lung = cancer,=20 particularly since the use of the anti-vascular agents, = which is=20 anticipated to increase the frequency of early detection of=20 otherwise asymptomatic CNS metastases for patients with = metastatic=20 disease.

CNS Metastases in = Patients=20 With Non-Small Cell Lung Cancer

As many as 50% to 60% of patients = with locally=20 advanced non-small cell lung cancer (NSCLC) will develop CNS = metastases during the course of their disease,16, = 70 = and the CNS is=20 the first site of recurrence in up to one third of all = patients=20 treated with multimodality treatment for stage III NSCLC (Table 4).70, = 71, = 72, = 73, = 74, = 75, = 76

Table 4.

Central Nervous System = Metastases=20 as First Site of Failure in Advanced Non-Small Cell = Lung=20 Cancer After Multimodality=20 Therapy


First Author, Year	No. of Patients	Patients With CNS Metastases as First = Site of=20 Relapse (%)	Percent of Total Relapses
Furuse, 199971	147	19
Law, 200172	42	29
Robnett, 200173	150	19
Ceresoli, 200274	139	22	29
Machtay, 200475	53	23
Gaspar, 200576	422	17	26
Mamon, = 200570	177	34

Abbreviation: CNS, = central=20 nervous = system.

The treatment of patients with NSCLC = has=20 developed substantially over the last 5 years. A large = percentage of=20 stage I-III patients with NSCLC are treated with adjuvant=20 chemotherapy. The Radiation Therapy Oncology Group 0139 = study also=20 showed that surgery in patients with stage III NSCLC does = not affect=20 the outcome for those who receive chemotherapy and radiation = first.=20 Furthermore, new drugs, such as bevacizumab and erlotinib = for=20 patients with stage IV lung cancer, have extended survival = and=20 substantially changed the clinician's practice.

Patients with advanced NSCLC remain = difficult=20 to treat and rarely attain a complete response. Conventional = platinum-based chemotherapy, including cisplatin plus = paclitaxel,=20 gemcitabine or docetaxel, and carboplatin plus paclitaxel, = provides=20 only modest survival benefits in patients with advanced = NSCLC=20 (median survival, 8 months; 1-year survival rate, = =E2=88=BC30%).77 = Several=20 recently developed therapies, including the antiangiogenic = drug=20 bevacizumab and the orally administered EGFR inhibitors = erlotinib=20 and gefitinib, have improved the outcome of patients with = advanced=20 NSCLC. The addition of bevacizumab to standard chemotherapy = regimens=20 has increased median survival for metastatic non-squamous = NSCLC to=20 greater than 12 months.69 = The effect of=20 bevacizumab treatment on the incidence of CNS metastases is=20 unclear.

In a recent randomized, = placebo-controlled=20 study (N =3D 731), erlotinib treatment resulted in a = response rate of=20 8.9% among previously treated patients with stage IIIb or = stage IV=20 NSCLC.78 = Overall=20 survival was 6.7 months in the erlotinib group, versus 4.7 = months in=20 the placebo group (P <.001). In patients with = advanced=20 NSCLC who failed to respond to conventional therapy, = gefitinib=20 produced response rates ranging from 12% to 18% and a 1-year = survival rate of 25%.16, = 79, = 80 = Despite the=20 ability to control systemic cancer, a recent study = demonstrated that=20 patients who respond to gefitinib treatment might be at = greater risk=20 of developing CNS relapse. Of 21 responders, seven (33%) had = CNS=20 metastases as first site of disease progression, with a = 5-year=20 actuarial incidence of CNS metastases of 60%.16 Of = note, two=20 (29%) patients with CNS relapses presented with NM. The = median=20 survival for patients with CNS disease treated with = gefitinib was 15=20 months, compared with 24 months for patients without CNS = disease.16 = Interestingly,=20 some reports have shown that gefitinib may have antitumor = activity=20 against CNS metastases from NSCLC at approximately the same=20 frequency as its systemic efficacy.74, = 81, = 82 = For example,=20 43% of patients with CNS metastases (n =3D 14) who had = received=20 gefitinib for advanced NSCLC achieved an objective response = in the=20 CNS, and 50% had extracranial responses.81 = Ceresoli and=20 colleagues prospectively evaluated 41 patients with NSCLC = and=20 measurable CNS metastases who were treated with gefitinib.74 A = partial=20 response in the brain occurred in four patients (10%), a = rate=20 observed in most trials of NSCLC with extracranial = disease.74

Subgroups of patients who are = potential=20 candidates for CNS prophylaxis could be defined by response = to=20 systemic therapy. For example, the likelihood of a response = to=20 erlotinib in previously treated patients with NSCLC has been = shown=20 to be significantly higher in women (P =3D .006), = nonsmokers=20 (P <.001), Asians (P =3D .02), patients = with=20 adenocarcinoma (P <.001), and those in whom EGFR = was=20 expressed in =E2=89=A510% of the tumor.78 = However, the=20 only factors that remained significant in multiple=20 logistic-regression analyses were nonsmokers, adenocarcinoma = histology, and EGFR expression.78

Identifying = High-Risk Patients=20 Who May Be Candidates for CNS Prophylactic=20 Treatment

By combining multiple factors, it = may be=20 possible to select a population with a high risk of CNS = disease to=20 justify CNS prophylaxis. For example, a clinical study could = be=20 conducted to evaluate the efficacy of prophylactic cranial=20 irradiation versus observation in patients on second- or = third-line=20 treatment (whose primary tumors are slow-growing) or with = low=20 standardized uptake value. A high standardized uptake value = is a=20 poor prognostic factor for survival in NSCLC.83 = Another=20 potential study could assess the efficacy of liposomal = cytarabine=20 versus observation in stage III patients in complete = remission who=20 previously received prophylactic cranial irradiation. = Alternately,=20 one could combine these strategies and consider a trial of=20 previously treated patients with metastatic NSCLC who have = responded=20 to an EGFR inhibitor and treat them with liposomal = cytarabine versus=20 observation as a prevention strategy for NM. Finally, = high-risk=20 patients who have been identified by CSF screening or other = means=20 for the presence of micro-metastasis could also be = considered for=20 treatment with liposomal cytarabine (positive CSF) or = observation=20 (negative CSF).

CNS Metastases in = Patients=20 With SCLC

Approximately 14% to 18% of patients = with SCLC=20 have verifiable CNS metastasis (parenchymal and NM) at the = initial=20 presentation.84, = 85 = After 2 years,=20 the cumulative risk of developing CNS metastases is almost = 50% in=20 patients with limited disease and 65% to 80% for those with=20 extensive disease.86 = The majority=20 of CNS metastases are located in the parenchyma, but NM is = not=20 uncommon and occurs in 8% to 18% of all patients with = SCLC.85, = 86, = 87, = 88, = 89 In = a=20 prospective study of 458 patients with SCLC, 70% of whom had = extensive disease, the prevalence of NM was 2% at diagnosis = and 10%=20 at the 2-year follow-up.15 = Most patients=20 with NM had metastases at other CNS locations. Treatment of = NM with=20 radiotherapy (if bulky disease, n =3D 9), systemic = chemotherapy (n =3D=20 5), or intra-CSF MTX (n =3D 4) did not influence survival.15 = Because CNS=20 metastases are more common in patients with SCLC, a CNS = prevention=20 paradigm can be established, and prophylactic cranial = irradiation is=20 recommended for patients with limited disease who have = achieved=20 complete response. Prophylactic cranial irradiation has been = shown=20 to improve overall survival and quality-adjusted life = expectancy90, = 91 = and reduces=20 the risk of CNS metastases by approximately 50%.90 = Similarly, it=20 improves disease-free survival and reduces the incidence of=20 symptomatic metastases in patients with advanced SCLC.92=20 Notwithstanding the clear benefits of prophylactic cranial=20 irradiation, other modalities, including intra-CSF therapy, = need to=20 be evaluated for the prevention of this complication. For = example, a=20 potential clinical trial could be designed to assess the = efficacy of=20 intra-CSF chemotherapies such as MTX, cytarabine, or = liposomal=20 cytarabine versus no chemoprophylaxis for the prevention of = NM in=20 patients with SCLC. Such a trial could include patients with = limited=20 or advanced-stage cancer who are in remission and have = received=20 prophylactic cranial irradiation.

Postsurgical=20 Prophylaxis

There is evidence to suggest that = surgical=20 resection of bulky CNS tumors, particularly those in the = posterior=20 fossa, increases the risk of developing NM.9, 10, = 11, = 12 = While surgery=20 for supratentorial metastases only leads to a 2% to 17% = incidence of=20 NM, surgery for posterior fossa metastases leads to a 10% to = 67%=20 incidence of NM (Table 5).9, 10, = 11, = 12 NM = is often=20 the only site of recurrence.12 = This raises=20 the question as to whether the increased incidence of NM in=20 posterior fossa metastases is a consequence of surgery or = can be=20 explained otherwise.

Table 5.

Incidence (%) of = Neoplastic=20 Meningitis After Surgical Resection of Brain=20 Metastases


First Author, Year	No. of Patients With Surgical = Resection	Cases of NM n (%)
Norris, 19979	N=20 =3D 66 Posterior fossa metastases: n =3D = 11 Supratentorial metastases: n =3D=20 55	4(36) 1(2)
van der Ree,199912	N=20 =3D 27 Posterior fossa, n =3D=20 9 Supratentorial metastases, n=20 =3D18	6=E2=80=A0(67) 3(17)
Siomin, 200410	N=20 =3D 18 Posterior fossa metastases: n =3D = 18	9(50)
Suki, = 200811	N=20 =3D 260 Posterior fossa metastases: N =3D = 260	26(10)

Abbreviation: NM, = neoplastic=20 meningitis.

Two additional patients = died of=20 progressive neurological disease, and two patients = were lost=20 to follow-up.

=E2=80=A0

In five patients, NM was = the only=20 site of = recurrence.

As indicated above, the location of = the=20 lesions could by itself contribute to the development of NM. = In one=20 study, patients with CNS metastases located in the posterior = fossa=20 were found to be at a higher risk (15%) of developing NM = compared=20 with patients with CNS metastases located elsewhere (10%).15 = The increased=20 risk of NM after posterior fossa surgery may be due to close = association with subarachnoid space and cisterna magna = fossa.11, = 93 = However,=20 several reports suggest that the type of surgery of = posterior fossa=20 metastases may play an important role in NM.10, = 11 = According to=20 these studies, the incidence of NM is significantly higher = after=20 surgery than after stereotactic radiosurgery for the = treatment of=20 posterior fossa metastases.10, = 11 In = a report by=20 Suki and colleagues, 79% of cases of NM occurred after = surgical=20 resection of posterior fossa metastases and 19% after = stereotactic=20 radiosurgery (P =3D .09).11 = Several=20 mechanisms have been proposed that may account for the = higher=20 incidence of NM after surgery of posterior fossa metastases, = including contamination of the surgical field with tumor = cells=20 during surgery. Piecemeal resection also is associated with = a=20 significantly higher rate of NM compared with en bloc = resection=20 (relative risk =3D 2.45; P =3D .02).11 = Although the=20 minimal number of tumor cells in the CSF that would be = sufficient to=20 cause NM is unknown, preclinical data suggest that the = introduction=20 of just 3,000 tumor cells into the CSF can lead to NM.94 = Thus, a unique=20 opportunity exists to investigate the value of CNS = prophylaxis in=20 patients who undergo posterior fossa surgery of isolated = metastases=20 by using chemotherapeutic prophylaxis during the = perioperative=20 period. An analogy is the use of prophylactic antibiotics = for=20 surgical introduction of intracranial ventricular shunts.95 A = randomized=20 study to determine whether chemotherapy for CNS prophylaxis = of NM=20 will reduce the incidence of NM in patients undergoing = surgery for=20 posterior fossa metastasis is proposed (Figure 3). End points should include = imaging and=20 CSF cytology analyses, as well as clinical outcome measures = such as=20 recurrence, progression, or death. Other follow-up = monitoring=20 options, such as quantitative reverse transcription = polymerase chain=20 reaction, brain and spinal MRI, and immunohistochemistry = also may be=20 valuable. MTX, cytarabine, thiotepa, and liposomal = cytarabine could=20 be administered either intrathecally via lumbar puncture or = Ommaya=20 reservoir, or in the case of MTX and cytarabine, = systemically=20 through intravenous infusion at the time of incision. = Potentially=20 confounding factors, such as the chemoresistance of some = tumors,=20 abnormal CSF flow, and bulky deposits, should be closely = evaluated=20 before patient enrollment.

Figure=20 3. Proposal for a pilot clinical trial = assessing=20 whether chemoprophylaxis reduces the incidence of = neoplastic=20 meningitis in patients with posterior fossa central = nervous=20 system metastases who undergo surgical resection. CSF, = cerebrospinal fluid; EVD, extra vascular drainage; = IHC,=20 immunohistochemistry; MRI, magnetic resonance=20 imaging.

Conclusions

New and more effective treatment = modalities=20 have prolonged the remission and survival of patients with = solid=20 tumors, including breast and lung cancer. Because of these = improved=20 systemic outcomes, the relative significance of CNS = metastases,=20 including NM, has increased dramatically. CNS metastases are = sometimes the only site of recurrence and the site that most = threatens patients' QoL and the survival of otherwise = responsive or=20 stable patients. CNS prophylaxis in subpopulations at risk = of CNS=20 relapse and better screening and early diagnosis may improve = clinical outcome and survival. The identification and = validation of=20 biological markers to predict CNS relapse would allow = clinicians to=20 identify patients who may benefit from CNS prophylactic = treatment.=20 Clinical trials with clearly defined end points also are = needed to=20 test the efficacy and safety of intra-CSF and other CNS = prophylaxis=20 in high-risk populations. It is hoped that the proposals in = this=20 paper stimulate discussion among clinicians and researchers = and lead=20 to the development of protocols for the prevention of NM and = CNS=20 metastases from solid tumors.

References

1.=20 1Jemal A, Siegel R, = Ward E,=20 Hao Y, Xu J, Murray T, = et al. Cancer statistics, 2008. = CA Cancer = J=20 Clin. 2008;58:71=E2=80=9396. CrossRef=20

2.=20 2Barnholtz-Sloan JS, = Sloan AE,=20 Davis FG, Vigneau FD, Lai P, = Sawaya RE.=20 Incidence proportions of brain = metastases=20 in patients diagnosed (1973 to 2001) in the Metropolitan = Detroit=20 Cancer Surveillance System. J Clin = Oncol.=20 2004;22:2865=E2=80=932872. CrossRef=20

3.=20 3Gleissner G, = Chamberlain MC.=20 Neoplastic meningitis. = Lancet = Neurol.=20 2006;5:443=E2=80=93452. Ab= stract=20 | Fu= ll=20 Text | Full-Text=20 PDF (181 KB) | MEDLINE=20 | CrossRef=20

4.=20 4Balm M, Hammack J. = Leptomeningeal carcinomatosis = (Presenting=20 features and prognostic factors). Arch = Neurol.=20 1996;53:626=E2=80=93632. MED= LINE

5.=20 5Bruno MK, Raizer J. = Leptomeningeal metastases from solid = tumors=20 (meningeal carcinomatosis). Cancer = Treat=20 Res. 2005;125:31=E2=80=9352. MED= LINE

6.=20 6Taillibert S, = Laigle-Donadey F,=20 Chodkiewicz C, Sanson M, Hoang-Xuan K,=20 Delattre JY. Leptomeningeal=20 metastases from solid malignancy: a review. = J = Neurooncol.=20 2005;75:85=E2=80=9399. CrossRef=20

7.=20 7Olson ME, Chernik NL,=20 Posner JB. Infiltration of the=20 leptomeninges by systemic cancer (A clinical and pathologic=20 study). Arch Neurol. 1974;30:122=E2=80=93137. MED= LINE

8.=20 8Wasserstrom WR, Glass JP,=20 Posner JB. Diagnosis = and=20 treatment of leptomeningeal metastases from solid tumors: = experience=20 with 90 patients. Cancer. 1982;49:759=E2=80=93772.

9.=20 9Norris LK, Grossman SA,=20 Olivi A. Neoplastic = meningitis=20 following surgical resection of isolated cerebellar = metastasis: a=20 potentially preventable complication. J = Neurooncol.=20 1997;32:215=E2=80=93223. CrossRef=20

10.=20 10Siomin VE, Vogelbaum MA,=20 Kanner AA, Lee SY, Suh JH, = Barnett GH.=20 Posterior fossa metastases: risk = of=20 leptomeningeal disease when treated with stereotactic = radiosurgery=20 compared to surgery. J Neurooncol. 2004;67:115=E2=80=93121. CrossRef=20

11.=20 11Suki D, Abouassi H, = Patel AJ,=20 Sawaya R, Weinberg JS, Groves MD. = Comparative risk of leptomeningeal = disease after=20 resection or stereotactic radiosurgery for solid tumor = metastasis to=20 the posterior fossa. J Neurosurg. 2008;108:248=E2=80=93257. CrossRef=20

12.=20 12van der Ree TC, Dippel DW, = Avezaat CJ, Sillevis Smitt PA, Vecht CJ, van = den=20 Bent MJ. Leptomeningeal=20 metastasis after surgical resection of brain = metastases.=20 J Neurol=20 Neurosurg Psych. 1999;66:225=E2=80=93227.

13.=20 13Bendell JC, Domchek SM,=20 Burstein HJ, Harris L, Younger J, = Kuter I,=20 et al. Central = nervous=20 system metastases in women who receive trastuzumab-based = therapy for=20 metastatic breast carcinoma. Cancer. 2003;97:2972=E2=80=932977.

14.=20 14Clayton AJ, Danson S, = Jolly S,=20 Ryder WD, Burt PA, Stewart AL,=20 et al. Incidence of=20 cerebral metastases in patients treated with trastuzumab for = metastatic breast cancer. Br J = Cancer.=20 2004;91:639=E2=80=93643. MED= LINE

15.=20 15Seute T, Leffers P, ten=20 Velde GP, Twijnstra A. Leptomeningeal metastases from small = cell lung=20 carcinoma. Cancer. 2005;104:1700=E2=80=931705.

16.=20 16Omuro AM, Kris MG, = Miller VA,=20 Franceschi E, Shah N, Milton DT, et = al. High incidence of disease recurrence = in the=20 brain and leptomeninges in patients with nonsmall cell lung=20 carcinoma after response to gefitinib. Cancer. 2005;103:2344=E2=80=932348.

17.=20 17Patel RR, Mehta MP. = Targeted therapy for brain metastases: = improving=20 the therapeutic ratio. Clin Cancer Res. 2007;13:1675=E2=80=931683. ME= DLINE=20 | CrossRef=20

18.=20 18Andrews DW, Scott CB,=20 Sperduto PW, Flanders AE, Gaspar LE, = Schell MC,=20 et al. Whole = brain=20 radiation therapy with or without stereotactic radiosurgery = boost=20 for patients with one to three brain metastases: phase III = results=20 of the RTOG 9508 randomised trial. Lancet. 2004;363:1665=E2=80=931672. Ab= stract=20 | Fu= ll=20 Text | Full-Text=20 PDF (425 KB) | CrossRef=20

19.=20 19Soffietti R, Ruda R,=20 Trevisan E. Brain = metastases:=20 current management and new developments. Curr Opin=20 Oncol. 2008;20:676=E2=80=93684. CrossRef=20

20.=20 20Jayson GC, Howell A, = Harris M,=20 Morgenstern G, Chang J, Ryder WD. = Carcinomatous meningitis in patients = with breast=20 cancer (An aggressive disease variant). Cancer. 1994;74:3135=E2=80=933141.

21.=20 21Glantz MJ, Cole BF, = Recht L,=20 Akerley W, Mills P, Saris S,=20 et al. High-dose=20 intravenous methotrexate for patients with nonleukemic=20 leptomeningeal cancer: is intrathecal chemotherapy=20 necessary?. J Clin Oncol. 1998;16:1561=E2=80=931567.

22.=20 22Chamberlain MC. Neoplastic meningitis. = Neurologist.=20 2006;12:179=E2=80=93187. MED= LINE=20 | CrossRef=20

23.=20 23Yap HY, Yap BS, = Rasmussen S,=20 Levens ME, Hortobagyi GN, = Blumenschein GR.=20 Treatment for meningeal = carcinomatosis in=20 breast cancer. Cancer. 1982;50:219=E2=80=93222.

24.=20 24Ongerboer de Visser BW, = Somers R,=20 Nooyen WH, van Heerde P, Hart AA,=20 McVie JG. Intraventricular=20 methotrexate therapy of leptomeningeal metastasis from = breast=20 carcinoma. Neurology. 1983;33:1565=E2=80=931572. ME= DLINE

25.=20 25Hitchins RN, Bell DR, = Woods RL,=20 Levi JA. A = prospective=20 randomized trial of single-agent versus combination = chemotherapy in=20 meningeal carcinomatosis. J Clin = Oncol.=20 1987;5:1655=E2=80=931662.

26.=20 26Grossman SA, = Finkelstein DM,=20 Ruckdeschel JC, Trump DL, Moynihan T,=20 Ettinger DS. Randomized=20 prospective comparison of intraventricular methotrexate and = thiotepa=20 in patients with previously untreated neoplastic meningitis = (Eastern=20 Cooperative Oncology Group). J Clin = Oncol.=20 1993;11:561=E2=80=93569.

27.=20 27Glantz MJ, Jaeckle KA,=20 Chamberlain MC, Phuphanich S, Recht L,=20 Swinnen LJ, et al. A=20 randomized controlled trial comparing intrathecal = sustained-release=20 cytarabine (DepoCyt) to intrathecal methotrexate in patients = with=20 neoplastic meningitis from solid tumors. Clin = Cancer=20 Res. 1999;5:3394=E2=80=933402. MED= LINE

28.=20 28Jaeckle KA, Batchelor T,=20 O'Day SJ, Phuphanich S, New P, Lesser G, = et al. An open = label=20 trial of sustained-release cytarabine (DepoCyt) for the = intrathecal=20 treatment of solid tumor neoplastic meningitis. = J = Neurooncol.=20 2002;57:231=E2=80=93239. CrossRef=20

29.=20 29Kim DY, Lee KW, = Yun T,=20 Park SR, Jung JY, Kim DW,=20 et al. Comparison of=20 intrathecal chemotherapy for leptomeningeal carcinomatosis = of a=20 solid tumor: methotrexate alone versus methotrexate in = combination=20 with cytosine arabinoside and hydrocortisone. = Jpn J Clin = Oncol. 2003;33:608=E2=80=93612. MED= LINE=20 | CrossRef=20

30.=20 30Boogerd W, van den = Bent MJ,=20 Koehler PJ, Heimans JJ, van der Sande JJ,=20 Aaronson NK, et al. The relevance of intraventricular = chemotherapy=20 for leptomeningeal metastasis in breast cancer: a randomised = study. Eur J Cancer. 2004;40:2726=E2=80=932733. Abstract=20 | Full=20 Text | Full-Text=20 PDF (278 KB) | MED= LINE=20 | CrossRef=20

31.=20 31Bokstein F, Lossos A,=20 Siegal T. Leptomeningeal=20 metastases from solid tumors: a comparison of two = prospective series=20 treated with and without intra-cerebrospinal fluid=20 chemotherapy. Cancer. 1998;82:1756=E2=80=931763.

32.=20 32Orlando L, Curigliano G,=20 Colleoni M, Fazio N, Nole F, = Martinelli G,=20 et al. Intrathecal=20 chemotherapy in carcinomatous meningitis from breast = cancer.=20 Anticancer Res. 2002;22:3057=E2=80=933059. ME= DLINE

33.=20 33Gavrilovic IT, = Posner JB.=20 Brain metastases: epidemiology = and=20 pathophysiology. J Neurooncol. 2005;75:5=E2=80=9314. CrossRef=20

34.=20 34Patanaphan V, Salazar OM,=20 Risco R. Breast = cancer:=20 metastatic patterns and their prognosis. South Med = J.=20 1988;81:1109=E2=80=931112. ME= DLINE

35.=20 35Tsukada Y, Fouad A, = Pickren JW,=20 Lane WW. Central = nervous=20 system metastasis from breast carcinoma (Autopsy = study).=20 Cancer. 1983;52:2349=E2=80=932354.

36.=20 36Kesari S, = Batchelor TT. Leptomeningeal metastases. = Neurol = Clin.=20 2003;21:25=E2=80=9366. Full=20 Text | Full-Text=20 PDF (651 KB) | MEDLINE=20 | CrossRef=20

37.=20 37Jaeckle KA, Phuphanich S,=20 Bent MJ, Aiken R, Batchelor T, = Campbell T,=20 et al. Intrathecal=20 treatment of neoplastic meningitis due to breast cancer with = a=20 slow-release formulation of cytarabine. Br J = Cancer.=20 2001;84:157=E2=80=93163. MED= LINE=20 | CrossRef=20

38.=20 38Wang ML, Yung WK, = Royce ME,=20 Schomer DF, Theriault RL. Capecitabine for = 5-fluorouracil-resistant brain=20 metastases from breast cancer. Am J Clin=20 Oncol. 2001;24:421=E2=80=93424. MED= LINE=20 | CrossRef=20

39.=20 39Fabi A, Vidiri A, = Ferretti G,=20 Felici A, Papaldo P, Carlini P,=20 et al. Dramatic = regression of multiple brain metastases from breast cancer = with=20 capecitabine: another arrow at the bow?. Cancer = Invest.=20 2006;24:466=E2=80=93468. MED= LINE=20 | CrossRef=20

40.=20 40Tham YL, Hinckley L, = Teh BS,=20 Elledge R. Long-term = clinical=20 response in leptomeningeal metastases from breast cancer = treated=20 with capecitabine monotherapy: a case report. = Clin = Breast=20 Cancer. 2006;7:164=E2=80=93166. MEDL= INE=20 | CrossRef=20

41.=20 41Giglio P, = Tremont-Lukats IW,=20 Groves MD. Response = of=20 neoplastic meningitis from solid tumors to oral = capecitabine.=20 J=20 Neurooncol. 2003;65:167=E2=80=93172. CrossRef=20

42.=20 42Rogers LR, Remer SE,=20 Tejwani S. Durable = response of=20 breast cancer leptomeningeal metastasis to capecitabine=20 monotherapy. Neurooncology. 2004;6:63=E2=80=9364. MEDLI= NE=20 | CrossRef=20

43.=20 43Capecitabine [package insert]. = Nutley,=20 NJ: Roche Laboratories Inc; 2006;.

44.=20 44Blum JL, Jones SE, = Buzdar AU,=20 LoRusso PM, Kuter I, Vogel C,=20 et al. Multicenter phase=20 II study of capecitabine in paclitaxel-refractory metastatic = breast=20 cancer. J Clin Oncol. 1999;17:485=E2=80=93493.

45.=20 45Gaspar L, Scott C, = Rotman M,=20 Asbell S, Phillips T, Wasserman T,=20 et al. Recursive=20 partitioning analysis (RPA) of prognostic factors in three = Radiation=20 Therapy Oncology Group (RTOG) brain metastases = trials.=20 Int J=20 Radiat Oncol Biol Phys. 1997;37:745=E2=80=93751. Full-Text=20 PDF (763 KB) | MED= LINE=20 | CrossRef=20

46.=20 46Slimane K, Andre F, = Delaloge S,=20 Dunant A, Perez A, Grenier J,=20 et al. Risk = factors for=20 brain relapse in patients with metastatic breast = cancer.=20 Ann=20 Oncol. 2004;15:1640=E2=80=931644. ME= DLINE=20 | CrossRef=20

47.=20 47Gabos Z, Sinha R, = Hanson J,=20 Chauhan N, Hugh J, Mackey JR,=20 et al. Prognostic=20 significance of human epidermal growth factor receptor = positivity=20 for the development of brain metastasis after newly = diagnosed breast=20 cancer. J Clin Oncol. 2006;24:5658=E2=80=935663. CrossRef=20

48.=20 48Souglakos J, Vamvakas L,=20 Apostolaki S, Perraki M, Saridaki Z, = Kazakou I,=20 et al. Central = nervous=20 system relapse in patients with breast cancer is associated = with=20 advanced stages, with the presence of circulating occult = tumor cells=20 and with the HER2/neu status. Breast = Cancer=20 Res. 2006;8:R36. CrossRef=20

49.=20 49Tham YL, Sexton K, = Kramer R,=20 Hilsenbeck S, Elledge R. Primary breast cancer phenotypes = associated with=20 propensity for central nervous system metastases.=20 Cancer. 2006;107:696=E2=80=93704.

50.=20 50Hicks DG, Short SM,=20 Prescott NL, Tarr SM, Coleman KA, = Yoder BJ,=20 et al. Breast = cancers=20 with brain metastases are more likely to be estrogen = receptor=20 negative, express the basal cytokeratin CK5/6, and = overexpress HER2=20 or EGFR. Am J Surg Pathol. 2006;30:1097=E2=80=931104. ME= DLINE

51.=20 51Carey LA, Ewend MG, = Metzger R,=20 Sawyer L, Dees EC, Sartor CI,=20 et al. Central = nervous=20 system metastases in women after multimodality therapy for = high risk=20 breast cancer. Breast Cancer Res Treat. 2004;88:273=E2=80=93280. MED= LINE=20 | CrossRef=20

52.=20 52Evans AJ, James JJ,=20 Cornford EJ, Chan SY, Burrell HC, = Pinder SE,=20 et al. Brain = metastases=20 from breast cancer: identification of a high-risk = group.=20 Clin=20 Oncol (R Coll Radiol). 2004;16:345=E2=80=93349. MED= LINE=20 | CrossRef=20

53.=20 53Gonzalez-Angulo AM, = Cristofanilli M,=20 Strom EA, Buzdar AU, Kau SW, Broglio KR, = et al. Central = nervous=20 system metastases in patients with high-risk breast = carcinoma after=20 multimodality treatment. Cancer. 2004;101:1760=E2=80=931766.

54.=20 54Ryberg M, Nielsen D,=20 Osterlind K, Andersen PK, Skovsgaard T,=20 Dombernowsky P. Predictors of=20 central nervous system metastasis in patients with = metastatic breast=20 cancer (A competing risk analysis of 579 patients treated = with=20 epirubicin-based chemotherapy). Breast = Cancer Res=20 Treat. 2005;91:217=E2=80=93225. MED= LINE=20 | CrossRef=20

55.=20 55Burstein HJ, Lieberman G,=20 Slamon DJ, Winer EP, Klein P. Isolated central nervous system = metastases in=20 patients with HER2-overexpressing advanced breast cancer = treated=20 with first-line trastuzumab-based therapy. = Ann = Oncol.=20 2005;16:1772=E2=80=931777. ME= DLINE=20 | CrossRef=20

56.=20 56Luck AA, Evans AJ, = Green AR,=20 Rakha EA, Paish C, Ellis IO. The influence of basal phenotype on = the=20 metastatic pattern of breast cancer. Clin Oncol = (R Coll=20 Radiol). 2008;20:40=E2=80=9345. CrossRef=20

57.=20 57Nam BH, Kim SY, = Han HS,=20 Kwon Y, Lee KS, Kim TH, et al. Breast cancer subtypes and survival in = patients=20 with brain metastases. Breast Cancer Res. 2008;10:R20. CrossRef=20

58.=20 58Baselga J, Tripathy D,=20 Mendelsohn J, Baughman S, Benz CC, = Dantis L,=20 et al. Phase II = study of=20 weekly intravenous recombinant humanized anti-p185HER2 = monoclonal=20 antibody in patients with HER2/neu-overexpressing metastatic = breast=20 cancer. J Clin Oncol. 1996;14:737=E2=80=93744.

59.=20 59Slamon DJ, = Leyland-Jones B,=20 Shak S, Fuchs H, Paton V, Bajamonde A,=20 et al. Use of=20 chemotherapy plus a monoclonal antibody against HER2 for = metastatic=20 breast cancer that overexpresses HER2. N Engl J = Med.=20 2001;344:783=E2=80=93792. ME= DLINE=20 | CrossRef=20

60.=20 60Lower EE, Drosick DR, = Blau R,=20 Brennan L, Danneman W, Hawley DK. = Increased rate of brain metastasis = with=20 trastuzumab therapy not associated with impaired = survival.=20 Clin=20 Breast Cancer. 2003;4:114=E2=80=93119. MEDL= INE=20 | CrossRef=20

61.=20 61Van Pelt AE, Mohsin S,=20 Elledge RM, Hilsenbeck SG, Gutierrez MC,=20 Lucci A, et al. Neoadjuvant trastuzumab and docetaxel = in breast=20 cancer: preliminary results. Clin = Breast=20 Cancer. 2003;4:348=E2=80=93353. MEDL= INE=20 | CrossRef=20

62.=20 62Lai R, Dang CT, = Malkin MG,=20 Abrey LE. The risk = of central=20 nervous system metastases after trastuzumab therapy in = patients with=20 breast carcinoma. Cancer. 2004;101:810=E2=80=93816.

63.=20 63Shmueli E, Wigler N,=20 Inbar M. Central = nervous=20 system progression among patients with metastatic breast = cancer=20 responding to trastuzumab treatment. Eur J = Cancer.=20 2004;40:379=E2=80=93382. Abstract=20 | Full=20 Text | Full-Text=20 PDF (114 KB) | MEDL= INE=20 | CrossRef=20

64.=20 64Stemmler HJ, Kahlert S,=20 Siekiera W, Untch M, Heinrich B,=20 Heinemann V. Characteristics=20 of patients with brain metastases receiving trastuzumab for = HER2=20 overexpressing metastatic breast cancer. Breast. 2006;15:219=E2=80=93225. Abstract= =20 | Full=20 Text | Full-Text=20 PDF (169 KB) | MEDLINE

65.=20 65Yau T, Swanton C, = Chua S,=20 Sue A, Walsh G, Rostom A,=20 et al. Incidence, pattern=20 and timing of brain metastases among patients with advanced = breast=20 cancer treated with trastuzumab. Acta = Oncol.=20 2006;45:196=E2=80=93201.

66.=20 66Geyer CE, Forster J,=20 Lindquist D, Chan S, Romieu CG, = Pienkowski T,=20 et al. Lapatinib plus=20 capecitabine for HER2-positive advanced breast = cancer.=20 N Engl J=20 Med. 2006;355:2733=E2=80=932743. CrossRef=20

67.=20 67Cameron D, Casey M, = Press M,=20 Lindquist D, Pienkowski T, Romieu CG,=20 et al. A phase = III=20 randomized comparison of lapatinib plus capecitabine versus=20 capecitabine alone in women with advanced breast cancer that = has=20 progressed on trastuzumab: updated efficacy and biomarker=20 analyses. Breast Cancer Res Treat. 2008;112:533=E2=80=93543. CrossRef=20

68.=20 68Lin NU, Carey LA, = Liu MC,=20 Younger J, Come SE, Ewend M,=20 et al. Phase II = trial of=20 lapatinib for brain metastases in patients with human = epidermal=20 growth factor receptor 2-positive breast cancer. = J Clin = Oncol.=20 2008;26:1993=E2=80=931999. CrossRef=20

69.=20 69Sandler A, Gray R, = Perry MC,=20 Brahmer J, Schiller JH, Dowlati A,=20 et al. Paclitaxel-carboplatin alone or with = bevacizumab=20 for non-small-cell lung cancer. N Engl J = Med.=20 2006;355:2542=E2=80=932550. CrossRef=20

70.=20 70Mamon HJ, Yeap BY, = Janne PA,=20 Reblando J, Shrager S, Jaklitsch MT,=20 et al. High = risk of brain=20 metastases in surgically staged IIIA non-small-cell lung = cancer=20 patients treated with surgery, chemotherapy, and = radiation.=20 J Clin=20 Oncol. 2005;23:1530=E2=80=931537. CrossRef=20

71.=20 71Furuse K, Fukuoka M,=20 Kawahara M, Nishikawa H, Takada Y, = Kudoh S,=20 et al. Phase = III study of=20 concurrent versus sequential thoracic radiotherapy in = combination=20 with mitomycin, vindesine, and cisplatin in unresectable = stage III=20 non-small-cell lung cancer. J Clin = Oncol.=20 1999;17:2692=E2=80=932699.

72.=20 72Law A, Karp DD, = Dipetrillo T,=20 Daly BT. Emergence = of=20 increased cerebral metastasis after high-dose preoperative=20 radiotherapy with chemotherapy in patients with locally = advanced=20 nonsmall cell lung carcinoma. Cancer. 2001;92:160=E2=80=93164.

73.=20 73Robnett TJ, Machtay M,=20 Stevenson JP, Algazy KM, Hahn SM. = Factors affecting the risk of brain = metastases=20 after definitive chemoradiation for locally advanced = non-small-cell=20 lung carcinoma. J Clin Oncol. 2001;19:1344=E2=80=931349.

74.=20 74Ceresoli GL, Reni M, = Chiesa G,=20 Carretta A, Schipani S, Passoni P,=20 et al. Brain = metastases=20 in locally advanced nonsmall cell lung carcinoma after = multimodality=20 treatment: risk factors analysis. Cancer. 2002;95:605=E2=80=93612.

75.=20 75Machtay M, Lee JH,=20 Stevenson JP, Shrager JB, Algazy KM, = Treat J, et=20 al. Two commonly used = neoadjuvant=20 chemoradiotherapy regimens for locally advanced stage III = non-small=20 cell lung carcinoma: long-term results and associations with = pathologic response. J Thorac Cardiovasc Surg. 2004;127:108=E2=80=93113. Abstract=20 | Full=20 Text | Full-Text=20 PDF (76 KB) | M= EDLINE=20 | CrossRef=20

76.=20 76Gaspar LE, Chansky K,=20 Albain KS, Vallieres E, Rusch V, = Crowley JJ,=20 et al. Time = from=20 treatment to subsequent diagnosis of brain metastases in = stage III=20 non-small-cell lung cancer: a retrospective review by the = Southwest=20 Oncology Group. J Clin Oncol. 2005;23:2955=E2=80=932961. CrossRef=20

77.=20 77Schiller JH, Harrington D, = Belani CP, Langer C, Sandler A, Krook J, = et al. Comparison of four=20 chemotherapy regimens for advanced non-small-cell lung=20 cancer. N Engl J Med. 2002;346:92=E2=80=9398. CrossRef=20

78.=20 78Shepherd FA, Rodrigues = Pereira J,=20 Ciuleanu T, Tan EH, Hirsh V, = Thongprasert S,=20 et al. Erlotinib in=20 previously treated non-small-cell lung cancer. = N Engl J = Med.=20 2005;353:123=E2=80=93132. CrossRef=20

79.=20 79Kris MG, Natale RB, = Herbst RS,=20 Lynch TJ, Prager D, Belani CP,=20 et al. Efficacy = of=20 gefitinib, an inhibitor of the epidermal growth factor = receptor=20 tyrosine kinase, in symptomatic patients with non-small cell = lung=20 cancer: a randomized trial. JAMA. 2003;290:2149=E2=80=932158. CrossRef=20

80.=20 80Fukuoka M, Yano S, = Giaccone G,=20 Tamura T, Nakagawa K, Douillard JY,=20 et al. Multi-institutional randomized phase = II trial of=20 gefitinib for previously treated patients with advanced=20 non-small-cell lung cancer (The IDEAL 1 Trial) = [corrected].=20 J Clin=20 Oncol. 2003;21:2237=E2=80=932246. CrossRef=20

81.=20 81Hotta K, Kiura K, = Ueoka H,=20 Tabata M, Fujiwara K, Kozuki T,=20 et al. Effect = of=20 gefitinib (=E2=80=98Iressa=E2=80=99, ZD1839) on brain = metastases in patients with=20 advanced non-small-cell lung cancer. Lung = Cancer.=20 2004;46:255=E2=80=93261. Abstract<= /A>=20 | Full=20 Text | Full-Text=20 PDF (86 KB) | MEDLINE=20 | CrossRef=20

82.=20 82Shimato S, Mitsudomi T,=20 Kosaka T, Yatabe Y, Wakabayashi T, = Mizuno M,=20 et al. EGFR = mutations in=20 patients with brain metastases from lung cancer: association = with=20 the efficacy of gefitinib. Neurooncol.=20 2006;8:137=E2=80=93144. MEDL= INE=20 | CrossRef=20

83.=20 83Berghmans T, Dusart M,=20 Paesmans M, Hossein-Foucher C, Buvat I,=20 Castaigne C, et al. Primary tumor standardized uptake = value (SUVmax)=20 measured on fluorodeoxyglucose positron emission tomography=20 (FDG-PET) is of prognostic value for survival in non-small = cell lung=20 cancer (NSCLC): a systematic review and meta-analysis (MA) = by the=20 European Lung Cancer Working Party for the IASLC Lung Cancer = Staging=20 Project. J Thorac Oncol. 2008;3:6=E2=80=9312.

84.=20 84Hochstenbag MM, = Twijnstra A,=20 Wilmink JT, Wouters EF, ten Velde GP. = Asymptomatic brain metastases (BM) in = small cell=20 lung cancer (SCLC): MR-imaging is useful at initial=20 diagnosis. J Neurooncol. 2000;48:243=E2=80=93248. CrossRef=20

85.=20 85Seute T, Leffers P, ten=20 Velde GP, Twijnstra A. Neurologic disorders in 432 = consecutive patients=20 with small cell lung carcinoma. Cancer. 2004;100:801=E2=80=93806.

86.=20 86Nugent JL, Bunn PA,=20 Matthews MJ, Ihde DC, Cohen MH, = Gazdar A,=20 et al. CNS = metastases in=20 small cell bronchogenic carcinoma: increasing frequency and = changing=20 pattern with lengthening survival. Cancer. 1979;44:1885=E2=80=931893.

87.=20 87Rosen ST, Aisner J, = Makuch RW,=20 Matthews MJ, Ihde DC, Whitacre M,=20 et al. Carcinomatous=20 leptomeningitis in small cell lung cancer: a = clinicopathologic=20 review of the National Cancer Institute experience.=20 Medicine. 1982;61:45=E2=80=9353.

88.=20 88Aisner J, Ostrow S, = Govindan S,=20 Wiernik PH. Leptomeningeal=20 carcinomatosis in small cell carcinoma of the lung.=20 Med=20 Pediatr Oncol. 1981;9:47=E2=80=9359. MEDLI= NE=20 | CrossRef=20

89.=20 89Aroney RS, Dalley DN, = Chan WK,=20 Bell DR, Levi JA. Meningeal carcinomatosis in small cell = carcinoma=20 of the lung. Am J Med. 1981;71:26=E2=80=9332. Abstract=20 | Full-Text=20 PDF (1716 KB) | MEDLINE<= /A>=20 | CrossRef=20

90.=20 90Auperin A, Arriagada R,=20 Pignon JP, Le Pechoux C, Gregor A, = Stephens RJ,=20 et al. Prophylactic=20 cranial irradiation for patients with small-cell lung cancer = in=20 complete remission (Prophylactic Cranial Irradiation = Overview=20 Collaborative Group). N Engl J Med. 1999;341:476=E2=80=93484. ME= DLINE=20 | CrossRef=20

91.=20 91Lee JJ, Bekele BN, = Zhou X,=20 Cantor SB, Komaki R, Lee JS. Decision analysis for prophylactic = cranial=20 irradiation for patients with small-cell lung cancer. = J Clin=20 Oncol. 2006;24:3597=E2=80=933603. CrossRef=20

92.=20 92Slotman B, Faivre-Finn C,=20 Kramer G, Rankin E, Snee M, Hatton M,=20 et al. Prophylactic=20 cranial irradiation in extensive small-cell lung = cancer.=20 N Engl J=20 Med. 2007;357:664=E2=80=93672. CrossRef=20

93.=20 93Mirimanoff RO, = Choi NC. Intradural spinal metastases in = patients with=20 posterior fossa brain metastases from various primary=20 cancers. Oncology. 1987;44:232=E2=80=93236.

94.=20 94Phillips PC, Than TT, = Cork LC,=20 Hilton J, Carson BS, Colvin OM, et al. = Intrathecal = 4-hydroperoxycyclophosphamide:=20 neurotoxicity, cerebrospinal fluid pharmacokinetics, and = antitumor=20 activity in a rabbit model of VX2 leptomeningeal=20 carcinomatosis. Cancer Res. 1992;52:6168=E2=80=936174. ME= DLINE

95.=20 95Ratilal R, Costa J,=20 Sampaio C. Antibiotic=20 prophylaxis for surgical introduction of intracranial = ventricular=20 shunts. Cochrane Database Syst Rev = (Online).=20 2006;3:CD005365. =

a Departments of Neuroscience and = Oncology,=20 University and San Giovanni Battista Hospital, Torino,=20 Italy

b Department of Internal Medicine, = Division of=20 Medical Oncology, Huntsman Cancer Institute, University of = Utah,=20 Salt Lake City, UT

c Department of Neurosurgery, Huntsman = Cancer=20 Institute, University of Utah Health Science Center, Salt = Lake City,=20 UT

d Otto-von-Guericke-University, = University=20 Hospital for Gynecology and Obstetrics, Magdeburg,=20 Germany

e Department of Medical Oncology,=20 Charit=C3=A9-Universitaetsmedizin Berlin, Berlin,=20 Germany

Address = correspondence=20 to Riccardo Soffietti, MD, Division of Neuro-Oncology, = University of=20 Torino and San Giovanni Battista Hospital, 15 Via Cherasco, = Torino,=20 10126, Italy

STATEMENT OF CONFLICT OF = INTEREST:=20 The authors have nothing to disclose.

PII: S0093-7754(09)00097-9

doi:10.1053/j.seminoncol.2009.05.007

3D""=20
9 of 10