- Email: [email protected]
Meningeal carcinomatosis in small cell carcinoma of the lung
Meningeal Carcinomatosis in Small Cell Carcinoma of the Lung
R. S. ARONEY, M.B., B.S. D. N. DALLEY, M.B., B.S.* W. K. CHAN, MB.,
B.S.
D. R. BELL, M.B., B.S. J. A. LEVI, M.B., B.S., F.R.A.C.P. St. Leonards, N.S. W. Australia
From the Department of Oncology, Royal North Shore Hospital of Sydney, St. Leonards, N.S.W. Requests for reprints should be addressed to Dr. R. S. Aroney, Department of Oncology, Royal North Shore Hospital of Sydney, St. Leonards, N.S.W. 2065, Australia. Manuscript accepted January 5, 1981. Present address: Department of Medical Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada. l
26
July 1981
Cerebral and meningeal metastases are increasingly important complications in small cell anaplastic carcinoma of the lung. In a study at this institution, 60 evaluable patients received intensive chemotherapy without prophylactic cranial irradiation or other prophylactic measures. The complete plus partial remission rate was 76 percent with a median survival of 49-l- weeks (range eight to 106+ weeks) for those with a complete response and 16+ weeks (range six to 67 weeks) for those with a partial response, all of which are comparable to other reported series. In 11 patients (16 percent) meningeal carcinomatosis has developed. Forty-two percent of the patients with a relapse have exhibited meningeal carcinomatosis and in 27 percent of the patients with a relapse it was the only site of relapse. Cerebral metastases occurred in 27 percent of those who had a relapse, and in 12 percent this was the sole site of relapse. Simultaneous meningeal carcinomatosis and cerebral metastases occurred in 6 percent of the patients with a relapse. The median time to meningeal relapse was 27 weeks (range 12 to 60 weeks) compared with 25+ weeks (six to 106+ weeks) over-all, and the median survival was 26 weeks (range 14 to 62 weeks) compared with 25-t weeks (two to 106+ weeks) for the whole group with small cell carcinoma of the lung. Meningeal involvement in small cell carcinoma of the lung must now be considered a sanctuary site of equal importance to cerebral metastases. To prevent and treat this complication will necessitate evaluation of all available modalities, including cranial and spinal irradiation, intrathecal chemotherapy and systemic agents that readily cross the blood-brain barrier. Small cell carcinoma of the lung has a marked propensity for early dissemination with the consequence that neither surgery nor radiotherapy alone have proved of great benefit to patients with this disease [ 11. Combination chemotherapy has become a principal component of the initial management of this condition because of its ability to induce major responses it? the majority of patients, with concomitant improvement in survival and diminution of symptoms in patients with both limited and extensive disease [2]. Unfortunately, most patients ultimately have a relapse at the sites of dominant disease pretreatment and/or in sanctuary sites [ 2,3]. It is the latter group that is the particular interest of this paper. In postmortem studies in the 1960s and early 1970s cerebral metastases was reported in 28 to 55 percent of the patients with small cell carcinoma of the lung [4], and although meningeal carcinomatosis was uncommonly found in early autopsy studies, in a more recent study it was found in 28 percent of the patients [5]. Prophylactic cranial irradiation fairly consistently reduces the incidence of cerebral metastases [6,7] but to date it has not
The American Journal of Medicine
Volume 71
MENINGEAL
CARCINOMATOSIS
convincingly improved survival because of relapse at other sites [8,9]. Furthermore, despite such prophylaxis, about 10 percent of the patients with extensive disease have been reported to have a relapse at central nervous system sites other than the brain, including meninges, dura and extradural sites [ lo]. Despite its growing importance, relatively little attention has been paid to the meningeal complications of small cell carcinoma of the lung [ 111. In a study at this institution 60 evaluable patients with small cell carcinoma of the lung received combination chemotherapy without central nervous system prophylaxis. We report here an unusually high over-all incidence of meningeal carcinomatosis in patients with relapse and most particularly draw attention to meninges as being the predominant if not the sole relapse site in many of these patients irrespective of the extent of their initial disease. When meningeal carcinomatosis occurred it was invariably the major determinant of subsequent prognosis, with few patients deriving enduring benefit from the therapies utilized. PATIENTS AND METHODS From mid 1977 to late 1979, 11 of 60 patients with small cell carcinoma of the lung were diagnosed as having meningeal carcinomatosis based on positive cerebrospinal fluid cytologic examination. The pretreatment characteristics of the total group with small cell carcinoma of the lung and of the patients in whom meningeal carcinomatosis subsequently developed are shown in Table I. Limited disease was defined as disease confined to the primary site and hilar and/or mediastinal lymph nodes, and extensive disease was defined
TABLE I
Pretreatment Characteristics Meningeal Over-all % No. -
Valuable
patients
Age (yr) Median Range Male:Female Limited disease Extensive disease Metastatic sites’ Liver Bone Marrow Brain Performance status+ O-II III-IV Systemic therapy CMVZ CMV alternating with VP-At
RelapseGroup No. %
ia
60 64
11 61
36-74 2.5: 1 17 43
28 72
52-71 1:l 3 a
27 73
15 16 10 6
25 27 17 10
4 3 2 3
36 27 ia 27
37 23
62 38
6 5
55 45
10 50
17 a3
3 a
27 73
* Not all inclusive. + Eastern Cooperative Oncology Group criteria. t See ter.t.
IN SMALL CELL CARCINOMA
TABLE II
OF LUNG--ARONEY
ET AL.
Dlagnostlc Features of Meningeal Relapses
Symptoms and signs Back pain Urinary incontinence/retention Impaired consciousness Asymmetric parasthesiae/hyperesthesia Para-/Quadra-paresis Personality/mood change Generalized asymmetric weakness Cranial nerve palsy Headache Cerebrospinal fluid Cytology Protein (g/liter) Median Range Glucose (g/liter) Median Range Cerebral scans Cerebral CAT Positive Normal Nuclear isotope Negative Not donet
7 5 5 4 4 3 2 2 2
64 46 46 36 36 27
11
+
ia ia ia
5.0 0.6-19.8’ 2.7 10.4-5.7’
27 46
NOTE: i- = positive. Normal 0.15-0.50 g/liter. + Normal 2.5-5.5 mmol/liter. t See text. l
as disease beyond the mediastinum. The first 10 patients with small cell carcinoma of the lung received cyclophosphamide 1,000 mg/m* intravenously on day 1, methotrexate 30 n-g/m* intravenously on days 1 and 5, and vincristine 1.4 mg/m* (maximum 2 mg) intravenously on days 1 and 5; the administration of all agents was repeated every 21 days (CMV). The subsequent 50 patients received CMV therapy alternating every 21 days with adriamycin, 45 mg/m* intravenously on day 1, plus VP 16-213, 75 mg/m* intravenously on days 1, 3 and 5 (VP-A). No prophylactic craniospinal therapy was given. Radiotherapy was not a standardized part of the treatment although localized radiotherapy was utilized in specific situations. Presenting features and diagnostic investigations of patients with meningeal carcinomatosis are given in Table II. Fungal and bacterial infections were excluded by routine microbiologic methods in all patients. Three patients had cerebral computerized axial tomography (CAT) scans (American Science and Engineering Type 500 scanner) which were positive for meningeal disease (Figure 1). Of the two nuclear isotope scans (technetium 9grn pertechnetate) performed, neither showed any abnormalities, but an electroencephalogram performed in one of these patients showed diffuse abnormalities. One patient in whom a cerebral metastasis was demonstrated by nuclear scan, and who was treated with cranial radiation four months earlier, did not have a further scan at the onset of meningeal carcinomatosis.
July 1991
The American Journal of Medicine
Volume 71
27
MENINGEAL CARCINOMATOSIS
Figure 1.
TABLE III
IN SMALL CELL CARCINOMA OF LUNG-ARONEY
Top, normal scan. Bottom, CAT scan showing pattern of meningeal carcinomatosis one month later.
Results of Therapy for Total Patient Population Complete Remission No.
Results of Therapy Total Limited disease Extensive disease Time to relapse (wk)” Total Limited disease Extensive disease Duration of survival (wk)’ Total Limited disease Extensive disease l
28
20 8 12
Partial Remission %
No.
33 47 28
27 7 20
Progressive Disease 56
No.
%
45 41 47
13 2 11
22 12 25
... ... ...
45+ (8-106+) 52+ (8-106+) 44+ (28+-105)
18+ (6-67) 25+ (13+-31) 15 (6-67)
... ... ...
49+ (8-106+) 58+ (8-106+) 46+ (20+-105)
20+ (6-72+) 26+ (13+-51) 18+ (6-72+)
6 (l-33) 9 (6-12) 7 (l-33)
Figures given are the median, those in parentheses the range.
July 1981
ET AL.
The American Journal of Yedlclne
Volume 71
MENINGEAL
t.
CR. PR.
C.---c
PO.
t-l
CARCINOMATOSIS
IN SMALL CELL CARCINOMA
OF LUNG--ARONEY
ET AL.
25
p<0.c01,CR+PRvsPD p=O.CQl.CRvsPR
50
75
0 0
22
44
66
66
110
100
Figure 2. Log rank comparison of survival curves in small cell carcinoma of the lung. CR = complete remission; PR = partial remission; PD = progressive disease.
Once the diagnosis of meningeal carcinomatosis was established, treatment consisted of the intrathecal administration of methotrexate (MTX), 15 mg in 3 ml Elliott’s 6 solution [ 12,131, by lumbar puncture thrice weekly for three weeks, twice weekly for two weeks and thereafter once every three or four weeks, depending upon adequacy of response. To avoid systemic effects of intrathecal MTX, 7.5 mg of folinic acid was given every 6 hours for four doses from the time of lumbar puncture. If progressive meningeal carcinomatosis occurred, intrathecal cytosine arabinoside (ara C), 50 mg/m’ in 3 ml of Elliott’s B solution, was added. All patients with meningeal carcinomatosis were given concurrent whole brain irradiation unless they had previously been given it for cerebral metastases present at the time of diagnosis or at an earlier stage of relapse. One patient decided against specific therapy. All cytologic, biochemical and microbial evaluations were repeated with each lumbar puncture. A response to intrathecal therapy was defined as the consistent absence of previously demonstrated malignant cells in cerebrospinal fluid, together with a return to normal of previously abnormal cerebrospinal fluid glucose levels and concomitant symptomatic improvement plus reduction in the severity of pretreatment neurologic deficits and an absence of new deficits. As cerebrospinal fluid protein levels can vary due to both changes in disease and intrathecal cytotoxic therapy, a response did not necessitate the return of cerebrospinal fluid protein levels to normal. Over-all response to treatment in the 60 patients with small cell carcinoma of the lung was defined as complete remission when there was disappearance of all clinically detectable tumor; partial remission when there was at least a 50 percent decrease in the size of the principal measurable or evaluable parameters of disease, with no new lesions appearing, and an associated improvement in performance status. Progressive disease was defined when the size of any preexisting measurable or evaluable lesion increased more than 25 percent, new lesions appeared or there was a significant deterioration in performance status. Evaluable parameters used to assess response included abnormalities on
22
0
weeks
44
bb
88
110
,
weeks
I-
1
Figure 3. Log rank comparison of times to relapse by response in small cell carcinoma of the lung. CR = complete remission; PI? = partial remission.
x-ray films, nucleotide and CAT scanning. For assessment of response to systemic chemotherapy, effects of radiation therapy on measurable parameters were excluded. RESULTS Of the 60 evaluable patients, the combined complete remission and partial remission rate for systemic disease was 78 percent, being 88 percent for patients with limited disease and 65 percent for patients with extensive disease (Table III). Taken from the time chemotherapy was started, the respective times to relapse and durations of survival are included in Table III and Figures 2 and 3; the over-all median time to relapse was 25i- weeks (6 to 106+ weeks) and the over-all survival median was 25-t weeks (2 to 106-I weeks). Of these 60 patients, 26 have had a relapse, and of these, 16 have exhibited central nervous system disease as a combined or sole relapse site (Table IV). For seven patients (27 percent of the relapses) meningeal
TABLE IV
Central Nervous System Relapses 41ol No.
Total Patients
% of Relapses
Total relapses (all sites) Total central nervous system relapses Meningeal relapses Alone Plus cerebral Plus other sites
26 16
43
100
27
62
7 2 2
12 3 3
27 6 8
Total
11
18
42
Cerebral relapses Alone Plus meningeal Plus other sites
3 2 2
5 3 3
12 6 8
Total
7
12
27
July 1991
The American Journal of Medicine
Volume 71
29
MENINGEAL CARCINOMATOSIS
TABLE V
IN SMALL CELL CARCINOMA
OF LUNG-ARONEY
ET AL.
Meningeal Relapses Complete Remission Median
Time to meningeal relapse Total Limited disease Extensive disease Duration of survival (wk) Total Limited disease Extensive disease
Partial Remission Range
50 52,60 46
24-60 24-50
58 58,82 53
75
a
3
50
-
If
-
-h-l
28-82 28-70
p = 0.28
25
3
22
44
4t 66
I
88
no
W&S
Log rank comparison of survival curves in smai Igure 4. cell carcinoma of the lung according to the occurrence of meningeal relapse.
30
July 1991
Median
Range
The Amertcan Journal of Medicine
18 31 18
21-31 12-30
.
21 41 21
15-41
14
15-38
h’
. .
from progressive disease after receiving only 1,000 rads. The median number of intrathecal MTX treatments was five, with a range of two to 15. Three patients unresponsive to intrathecal MTX received, respectively, one, five and nine intrathecal cytosine arabinoside (ara C) treatments, but no responses were seen. Durations of over-all response and subsequent survival were disappointingly short as shown in Table VI, with significantly better outcome, however, for those who responded, compared with those whose disease progressed (p <0.02; x2 6.27; 1 degree of freedom). All those who responded survived eight weeks or more whereas all those who did not respond died within six weeks. Toxicity. Over-all toxicity of meningeal carcinomatosis treatment was not great. Residual controllable lumbar ache was common after three lumbar punctures even in those patients with responsive disease. In one patient chemical meningism developed after two doses of MTX but this was responsive to intrathecal hydrocortisone, 200 mg, given together with intrathecal MTX in the subsequent 12 lumbar punctures. In only one patient (receiving intrathecal MTX) did myelosuppression attributable to intrathecal therapy develop. There was no other significant toxicity. COMMENTS
MellhgdRdapse NotvkmiweslRekwe
z
0
Range
(wk)
disease was the only site of relapse and for three patients (12 percent of the relapses) cerebral metastases was the sole site of relapse. Over-all, 42 percent of the patients with a relapse had meningeal carcinomatosis either alone or in conjunction with other sites, and similarly 27 percent had cerebral metastases. Not one pretreatment characteristic, including cerebral metastases at the time of diagnosis, had a statistically significant correlation with the later development of meningeal carcinomatosis. Patients manifesting meningeal carcinomatosis had response, time to relapse and survival characteristics which did not differ significantly from those in the over-all group with small cell carcinoma of the lung (Table V and Figure 4) or from those in whom meningeal carcinomatosis did not develop. In 10 of the 11 patients with meningeal carcinomatosis therapeutic control was attempted (Table VI) and in five patients (50 percent) worthwhile relief of symptoms and neurologic impairment was obtained. All 10 patients received intrathecal MTX therapy and four concurrently commenced whole brain irradiation, although only two patients completed the intended 4,000 rads in four weeks, the other two patients succumbed
100
Median
Disease Progressive
Volume 71
The first case report of leptomeningeal metastasis was probably that of Eberth in 1870 [ 141, and the association with systemic malignancy was first recognized by Benda in 1901 [ 151. Up to the early 197Os, case reports totalled only 250 for all malignancies. It is well recognized that the diagnosis may be missed even at autopsy because macroscopic changes may be slight, and the infiltration may be confined to the spinal meninges. The lack of localizing signs and the protean nature of the disease often leads the uninitiated clinician astray, but paradoxically it is the diversity of symptoms and multifocal findings that should alert one to the diagnosis. Patients often offer myriad complaints, sometimes minor or vague, and frequently they are only elicited by
MENINGEAL CARCINOMATOSIS
TABLE Vi
Meningeal
IN SMALL CELL CARCINOMA
OF LUNG-ARONEY
ET AL.
Relapses-Results Responders No.
Results of therapy lntrathecal methotrexate alone Plus cranial radiotherapy (R,T) Plus cranial R,T and cytosine arabinoside No therapy given Duration of response (wk) Median Duration of survival from time meningial carcinomatosis diagnosed (wk) Over-all Responders Nonresponders
direct questioning [ 161. The meninges and brain as pharmacologic sanctuary sites have recently become increasingly important in the management of several malignancies, particularly the leukemias and lymphomas. Small cell carcinoma of the lung is now also of particular relevance in this context as intensive chemotherapy has improved the prognosis of this subtype. The high response rates and prolonged survival patterns of our patients are comparable with the better combinations reported by other workers, but the extreme prominence of relapses within the central nervous system, especially meningeal, highlights the sanctuary site problem. Meningeal involvement in small cell carcinoma of the lung must now be considered a sanctuary site of equal importance to cerebral metastases, particularly as the outlook for patients from the time of relapse within the central nervous system is not good, and is particularly poor for those with meningeai carcinomatosis. Although gratifying and worthwhile control of symptoms and neuroiogic deficits was achieved in 50 percent of our patients with meningeal carcinomatosis, the longest response was less than two months and the best survival was only a little over four months (Table VI). Prevention would obviously be preferable. Our patients received neither prophylactic cranial or spinal irradiation, nor prophylactic intrathecai therapy. Furthermore, none of the chemotherapeutic agents in the doses used are known to achieve therapeutic concentrations in the central nervous system. Certainly, prophylactic cranial irradiation appears to reduce the incidence of cerebral mass lesion relapses, but it confers no definite survival advantage [6,7] and has not yet been evaluated for its possible influence on the development of spinal meningeal disease. Chemotherapeutic agents which readily cross the blood-brain barrier, such as l-(2-chloroethyi)-3-cyclohexyl-l-ni-
Range
No. 3 1 1
5 4 1 1 a
4-10
5 11 2
2-10 8-17 4-6
%60 25 100
trosourea (CCNU) and procarbazine have not been demonstrated to influence the incidence of central nervous system relapses [ 171, however, the ultimate role of these agents and of high dose MTX programs will require randomized prospective trials. Prophylactic spinal irradiation and/or intrathecal cytotoxic therapy remain essentially unevaluated, understandably because of the problems of myelosuppression and other toxicities particularly with concurrent systemic chemotherapy. It is not possible at this stage to determine the exact basis for the inordinately high incidence of meningeal carcinomatosis reported here. it would seem likely that more than one factor is operative. Firstly, the extended survival resulting from high levels of systemic disease control, allowing time for the progression of meningeal disease to the point of clinical detection. Even autopsy evaluations of therapy programs, less effective in extending survival than is now the rule, have repeatedly demonstrated central nervous system involvement in from 28 to 55 percent of the patients. Secondly is the absence of systemic chemotherapeutic agents with significant central nervous system penetration. Thirdly is the possibility that the cytotoxic agents in combination might actively predispose to the development of meningeai carcinomatosis by some as yet unknown mechanism. Finally, it is possible, that the incidence has been underreported to date. Certainly, attention to all possible mechanisms of meningeal carcinomatosis relapse will be required if effective prophylactic programs are to be designed. There is also a pressing need for more effective management of meningeal carcinomatosis relapse. Combination intrathecal cytotoxic therapy warrants further evaluation, but major advances will probably require the availability of new effective agents and/or better tolerated, yet more effective, radiotherapeutic approaches for control of leptomeningeal disease.
July 1991
The American Journal of Medicine
Volume 71
31
MENINGEAL CARCINOMATOSIS
1N SMALL CELL CARCINOMA OF LUNG-ARONEY
ET AL.
REFERENCES 1.
32
Roswit B, Patno ME, Rapp R. et al.: The survival of patients with inoperable lung cancer. A large-scale randomized study of radiotherapy versus placebo. Radiology 1966; 90: 686-697. Greco FA, Einhorn LH, Richardson RL, Okfham RK: Small cell lung cancer: progress and perspectives. Semin Oncol 1976; 5: 323-335. Einhorn LH, Bond WH, Hornback N, Beng-Tek J: Long-term results in combined-modality treatment of small cell carcinoma of the lung. Semin Oncol 1976; 5: 309-313. Bunn PA, Nugent JL, Matthews MJ: Central nervous system metastases in small cell bronchogenic carcinoma. Semin Oncol 1976; 5: 314-322. Nugent JL, Bunn PA, lhde DC, et al.: Diagnosis and treatment of CNS complications of small cell bronchogenic carcinoma (SCBC). Proceedings of the American Association for Cancer Research and the American Society of Clinical Oncology 1978; 19: 116. Jackson DV, Richards F, Cooper MR. et al: Prophylactic cranial irradiation in small cell carcinoma of the lung: a randomized study. JAMA 1977; 237: 2730-2733. Holoye PY, Sam&s ML. Lanzotti VJ, Smith T, Barkley HT Jr: Combination chemotherapy and radiation therapy for small cell carcinoma. JAMA 1977; 237: 1221-1224. Chan HC, Carey RW: Small cell anaplastic carcinoma of lung. Reappraisal of current management. Cancer 1976; 37: 2651-2657. Hornback NB, Einhorn L, Skidnia H, Joe BT, Krause M, Furnas B: Oat cell carcinoma of the lung. Early treatment results
July 1961
The American Journal of Medicine
Volume 71
10.
11.
12.
13. 14.
15.
16.
17.
of combination radiation therapy and chemotherapy. Cancer 1976; 37: 2658-2664. Matthews MJ, Gazdar AF, lhde DC, et al.: Histologic subtypes of small cell carcinoma of the lung and their clinical significance. Proceedings of the American Society of Clinical Oncology 1978; 19: 397. Brereton HC, O’Donnell JF, Kent CH. et al.: Spinal meningeal carcinomatosis in small-cell carcinoma of the lung. Ann Intern Med 1978; 88: 517-519. Elliott KAC, et al.: Physiological salt solutions for brain surgery. Studies of local pH and pial vessel reactions to buffered and unbuffered isotonic solutions. J Neurosurg 1949; 6: 140-152. Lewis RC, et al.: Clinical uses of an artificial cerebrospinal fluid. J Neurosurg 1950; 7: 256-260. Eberth CJ: Zur Entwickluna des Eoithetioms (Cholesteatoms) der Pia und der Lunge. Vi&how Arch ‘1870; 49: 51-63. Lilienfeld J, Benda C: Fall von Metastischer Karcinose der Neivenund Hirnhaute. Berlin Klin Wochenschr 1901; 38: 729-730. Olsen ME, Chernlk NL, Posner JB: Infiltration of the leptomeninges by systemic cancer. Arch Neurol 1974; 30: 122-137. Nugent JL, Bunn PA, lhde DC, et al.: Diagnosis and treatment of CNS complications of small cell bronchogenic carcinoma. Proceedings of the American Association for Cancer Research and the American Society of Clinical Oncology 1978; 19: 116.
R. S. ARONEY, M.B., B.S. D. N. DALLEY, M.B., B.S.* W. K. CHAN, MB.,
B.S.
D. R. BELL, M.B., B.S. J. A. LEVI, M.B., B.S., F.R.A.C.P. St. Leonards, N.S. W. Australia
From the Department of Oncology, Royal North Shore Hospital of Sydney, St. Leonards, N.S.W. Requests for reprints should be addressed to Dr. R. S. Aroney, Department of Oncology, Royal North Shore Hospital of Sydney, St. Leonards, N.S.W. 2065, Australia. Manuscript accepted January 5, 1981. Present address: Department of Medical Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada. l
26
July 1981
Cerebral and meningeal metastases are increasingly important complications in small cell anaplastic carcinoma of the lung. In a study at this institution, 60 evaluable patients received intensive chemotherapy without prophylactic cranial irradiation or other prophylactic measures. The complete plus partial remission rate was 76 percent with a median survival of 49-l- weeks (range eight to 106+ weeks) for those with a complete response and 16+ weeks (range six to 67 weeks) for those with a partial response, all of which are comparable to other reported series. In 11 patients (16 percent) meningeal carcinomatosis has developed. Forty-two percent of the patients with a relapse have exhibited meningeal carcinomatosis and in 27 percent of the patients with a relapse it was the only site of relapse. Cerebral metastases occurred in 27 percent of those who had a relapse, and in 12 percent this was the sole site of relapse. Simultaneous meningeal carcinomatosis and cerebral metastases occurred in 6 percent of the patients with a relapse. The median time to meningeal relapse was 27 weeks (range 12 to 60 weeks) compared with 25+ weeks (six to 106+ weeks) over-all, and the median survival was 26 weeks (range 14 to 62 weeks) compared with 25-t weeks (two to 106+ weeks) for the whole group with small cell carcinoma of the lung. Meningeal involvement in small cell carcinoma of the lung must now be considered a sanctuary site of equal importance to cerebral metastases. To prevent and treat this complication will necessitate evaluation of all available modalities, including cranial and spinal irradiation, intrathecal chemotherapy and systemic agents that readily cross the blood-brain barrier. Small cell carcinoma of the lung has a marked propensity for early dissemination with the consequence that neither surgery nor radiotherapy alone have proved of great benefit to patients with this disease [ 11. Combination chemotherapy has become a principal component of the initial management of this condition because of its ability to induce major responses it? the majority of patients, with concomitant improvement in survival and diminution of symptoms in patients with both limited and extensive disease [2]. Unfortunately, most patients ultimately have a relapse at the sites of dominant disease pretreatment and/or in sanctuary sites [ 2,3]. It is the latter group that is the particular interest of this paper. In postmortem studies in the 1960s and early 1970s cerebral metastases was reported in 28 to 55 percent of the patients with small cell carcinoma of the lung [4], and although meningeal carcinomatosis was uncommonly found in early autopsy studies, in a more recent study it was found in 28 percent of the patients [5]. Prophylactic cranial irradiation fairly consistently reduces the incidence of cerebral metastases [6,7] but to date it has not
The American Journal of Medicine
Volume 71
MENINGEAL
CARCINOMATOSIS
convincingly improved survival because of relapse at other sites [8,9]. Furthermore, despite such prophylaxis, about 10 percent of the patients with extensive disease have been reported to have a relapse at central nervous system sites other than the brain, including meninges, dura and extradural sites [ lo]. Despite its growing importance, relatively little attention has been paid to the meningeal complications of small cell carcinoma of the lung [ 111. In a study at this institution 60 evaluable patients with small cell carcinoma of the lung received combination chemotherapy without central nervous system prophylaxis. We report here an unusually high over-all incidence of meningeal carcinomatosis in patients with relapse and most particularly draw attention to meninges as being the predominant if not the sole relapse site in many of these patients irrespective of the extent of their initial disease. When meningeal carcinomatosis occurred it was invariably the major determinant of subsequent prognosis, with few patients deriving enduring benefit from the therapies utilized. PATIENTS AND METHODS From mid 1977 to late 1979, 11 of 60 patients with small cell carcinoma of the lung were diagnosed as having meningeal carcinomatosis based on positive cerebrospinal fluid cytologic examination. The pretreatment characteristics of the total group with small cell carcinoma of the lung and of the patients in whom meningeal carcinomatosis subsequently developed are shown in Table I. Limited disease was defined as disease confined to the primary site and hilar and/or mediastinal lymph nodes, and extensive disease was defined
TABLE I
Pretreatment Characteristics Meningeal Over-all % No. -
Valuable
patients
Age (yr) Median Range Male:Female Limited disease Extensive disease Metastatic sites’ Liver Bone Marrow Brain Performance status+ O-II III-IV Systemic therapy CMVZ CMV alternating with VP-At
RelapseGroup No. %
ia
60 64
11 61
36-74 2.5: 1 17 43
28 72
52-71 1:l 3 a
27 73
15 16 10 6
25 27 17 10
4 3 2 3
36 27 ia 27
37 23
62 38
6 5
55 45
10 50
17 a3
3 a
27 73
* Not all inclusive. + Eastern Cooperative Oncology Group criteria. t See ter.t.
IN SMALL CELL CARCINOMA
TABLE II
OF LUNG--ARONEY
ET AL.
Dlagnostlc Features of Meningeal Relapses
Symptoms and signs Back pain Urinary incontinence/retention Impaired consciousness Asymmetric parasthesiae/hyperesthesia Para-/Quadra-paresis Personality/mood change Generalized asymmetric weakness Cranial nerve palsy Headache Cerebrospinal fluid Cytology Protein (g/liter) Median Range Glucose (g/liter) Median Range Cerebral scans Cerebral CAT Positive Normal Nuclear isotope Negative Not donet
7 5 5 4 4 3 2 2 2
64 46 46 36 36 27
11
+
ia ia ia
5.0 0.6-19.8’ 2.7 10.4-5.7’
27 46
NOTE: i- = positive. Normal 0.15-0.50 g/liter. + Normal 2.5-5.5 mmol/liter. t See text. l
as disease beyond the mediastinum. The first 10 patients with small cell carcinoma of the lung received cyclophosphamide 1,000 mg/m* intravenously on day 1, methotrexate 30 n-g/m* intravenously on days 1 and 5, and vincristine 1.4 mg/m* (maximum 2 mg) intravenously on days 1 and 5; the administration of all agents was repeated every 21 days (CMV). The subsequent 50 patients received CMV therapy alternating every 21 days with adriamycin, 45 mg/m* intravenously on day 1, plus VP 16-213, 75 mg/m* intravenously on days 1, 3 and 5 (VP-A). No prophylactic craniospinal therapy was given. Radiotherapy was not a standardized part of the treatment although localized radiotherapy was utilized in specific situations. Presenting features and diagnostic investigations of patients with meningeal carcinomatosis are given in Table II. Fungal and bacterial infections were excluded by routine microbiologic methods in all patients. Three patients had cerebral computerized axial tomography (CAT) scans (American Science and Engineering Type 500 scanner) which were positive for meningeal disease (Figure 1). Of the two nuclear isotope scans (technetium 9grn pertechnetate) performed, neither showed any abnormalities, but an electroencephalogram performed in one of these patients showed diffuse abnormalities. One patient in whom a cerebral metastasis was demonstrated by nuclear scan, and who was treated with cranial radiation four months earlier, did not have a further scan at the onset of meningeal carcinomatosis.
July 1991
The American Journal of Medicine
Volume 71
27
MENINGEAL CARCINOMATOSIS
Figure 1.
TABLE III
IN SMALL CELL CARCINOMA OF LUNG-ARONEY
Top, normal scan. Bottom, CAT scan showing pattern of meningeal carcinomatosis one month later.
Results of Therapy for Total Patient Population Complete Remission No.
Results of Therapy Total Limited disease Extensive disease Time to relapse (wk)” Total Limited disease Extensive disease Duration of survival (wk)’ Total Limited disease Extensive disease l
28
20 8 12
Partial Remission %
No.
33 47 28
27 7 20
Progressive Disease 56
No.
%
45 41 47
13 2 11
22 12 25
... ... ...
45+ (8-106+) 52+ (8-106+) 44+ (28+-105)
18+ (6-67) 25+ (13+-31) 15 (6-67)
... ... ...
49+ (8-106+) 58+ (8-106+) 46+ (20+-105)
20+ (6-72+) 26+ (13+-51) 18+ (6-72+)
6 (l-33) 9 (6-12) 7 (l-33)
Figures given are the median, those in parentheses the range.
July 1981
ET AL.
The American Journal of Yedlclne
Volume 71
MENINGEAL
t.
CR. PR.
C.---c
PO.
t-l
CARCINOMATOSIS
IN SMALL CELL CARCINOMA
OF LUNG--ARONEY
ET AL.
25
p<0.c01,CR+PRvsPD p=O.CQl.CRvsPR
50
75
0 0
22
44
66
66
110
100
Figure 2. Log rank comparison of survival curves in small cell carcinoma of the lung. CR = complete remission; PR = partial remission; PD = progressive disease.
Once the diagnosis of meningeal carcinomatosis was established, treatment consisted of the intrathecal administration of methotrexate (MTX), 15 mg in 3 ml Elliott’s 6 solution [ 12,131, by lumbar puncture thrice weekly for three weeks, twice weekly for two weeks and thereafter once every three or four weeks, depending upon adequacy of response. To avoid systemic effects of intrathecal MTX, 7.5 mg of folinic acid was given every 6 hours for four doses from the time of lumbar puncture. If progressive meningeal carcinomatosis occurred, intrathecal cytosine arabinoside (ara C), 50 mg/m’ in 3 ml of Elliott’s B solution, was added. All patients with meningeal carcinomatosis were given concurrent whole brain irradiation unless they had previously been given it for cerebral metastases present at the time of diagnosis or at an earlier stage of relapse. One patient decided against specific therapy. All cytologic, biochemical and microbial evaluations were repeated with each lumbar puncture. A response to intrathecal therapy was defined as the consistent absence of previously demonstrated malignant cells in cerebrospinal fluid, together with a return to normal of previously abnormal cerebrospinal fluid glucose levels and concomitant symptomatic improvement plus reduction in the severity of pretreatment neurologic deficits and an absence of new deficits. As cerebrospinal fluid protein levels can vary due to both changes in disease and intrathecal cytotoxic therapy, a response did not necessitate the return of cerebrospinal fluid protein levels to normal. Over-all response to treatment in the 60 patients with small cell carcinoma of the lung was defined as complete remission when there was disappearance of all clinically detectable tumor; partial remission when there was at least a 50 percent decrease in the size of the principal measurable or evaluable parameters of disease, with no new lesions appearing, and an associated improvement in performance status. Progressive disease was defined when the size of any preexisting measurable or evaluable lesion increased more than 25 percent, new lesions appeared or there was a significant deterioration in performance status. Evaluable parameters used to assess response included abnormalities on
22
0
weeks
44
bb
88
110
,
weeks
I-
1
Figure 3. Log rank comparison of times to relapse by response in small cell carcinoma of the lung. CR = complete remission; PI? = partial remission.
x-ray films, nucleotide and CAT scanning. For assessment of response to systemic chemotherapy, effects of radiation therapy on measurable parameters were excluded. RESULTS Of the 60 evaluable patients, the combined complete remission and partial remission rate for systemic disease was 78 percent, being 88 percent for patients with limited disease and 65 percent for patients with extensive disease (Table III). Taken from the time chemotherapy was started, the respective times to relapse and durations of survival are included in Table III and Figures 2 and 3; the over-all median time to relapse was 25i- weeks (6 to 106+ weeks) and the over-all survival median was 25-t weeks (2 to 106-I weeks). Of these 60 patients, 26 have had a relapse, and of these, 16 have exhibited central nervous system disease as a combined or sole relapse site (Table IV). For seven patients (27 percent of the relapses) meningeal
TABLE IV
Central Nervous System Relapses 41ol No.
Total Patients
% of Relapses
Total relapses (all sites) Total central nervous system relapses Meningeal relapses Alone Plus cerebral Plus other sites
26 16
43
100
27
62
7 2 2
12 3 3
27 6 8
Total
11
18
42
Cerebral relapses Alone Plus meningeal Plus other sites
3 2 2
5 3 3
12 6 8
Total
7
12
27
July 1991
The American Journal of Medicine
Volume 71
29
MENINGEAL CARCINOMATOSIS
TABLE V
IN SMALL CELL CARCINOMA
OF LUNG-ARONEY
ET AL.
Meningeal Relapses Complete Remission Median
Time to meningeal relapse Total Limited disease Extensive disease Duration of survival (wk) Total Limited disease Extensive disease
Partial Remission Range
50 52,60 46
24-60 24-50
58 58,82 53
75
a
3
50
-
If
-
-h-l
28-82 28-70
p = 0.28
25
3
22
44
4t 66
I
88
no
W&S
Log rank comparison of survival curves in smai Igure 4. cell carcinoma of the lung according to the occurrence of meningeal relapse.
30
July 1991
Median
Range
The Amertcan Journal of Medicine
18 31 18
21-31 12-30
.
21 41 21
15-41
14
15-38
h’
. .
from progressive disease after receiving only 1,000 rads. The median number of intrathecal MTX treatments was five, with a range of two to 15. Three patients unresponsive to intrathecal MTX received, respectively, one, five and nine intrathecal cytosine arabinoside (ara C) treatments, but no responses were seen. Durations of over-all response and subsequent survival were disappointingly short as shown in Table VI, with significantly better outcome, however, for those who responded, compared with those whose disease progressed (p <0.02; x2 6.27; 1 degree of freedom). All those who responded survived eight weeks or more whereas all those who did not respond died within six weeks. Toxicity. Over-all toxicity of meningeal carcinomatosis treatment was not great. Residual controllable lumbar ache was common after three lumbar punctures even in those patients with responsive disease. In one patient chemical meningism developed after two doses of MTX but this was responsive to intrathecal hydrocortisone, 200 mg, given together with intrathecal MTX in the subsequent 12 lumbar punctures. In only one patient (receiving intrathecal MTX) did myelosuppression attributable to intrathecal therapy develop. There was no other significant toxicity. COMMENTS
MellhgdRdapse NotvkmiweslRekwe
z
0
Range
(wk)
disease was the only site of relapse and for three patients (12 percent of the relapses) cerebral metastases was the sole site of relapse. Over-all, 42 percent of the patients with a relapse had meningeal carcinomatosis either alone or in conjunction with other sites, and similarly 27 percent had cerebral metastases. Not one pretreatment characteristic, including cerebral metastases at the time of diagnosis, had a statistically significant correlation with the later development of meningeal carcinomatosis. Patients manifesting meningeal carcinomatosis had response, time to relapse and survival characteristics which did not differ significantly from those in the over-all group with small cell carcinoma of the lung (Table V and Figure 4) or from those in whom meningeal carcinomatosis did not develop. In 10 of the 11 patients with meningeal carcinomatosis therapeutic control was attempted (Table VI) and in five patients (50 percent) worthwhile relief of symptoms and neurologic impairment was obtained. All 10 patients received intrathecal MTX therapy and four concurrently commenced whole brain irradiation, although only two patients completed the intended 4,000 rads in four weeks, the other two patients succumbed
100
Median
Disease Progressive
Volume 71
The first case report of leptomeningeal metastasis was probably that of Eberth in 1870 [ 141, and the association with systemic malignancy was first recognized by Benda in 1901 [ 151. Up to the early 197Os, case reports totalled only 250 for all malignancies. It is well recognized that the diagnosis may be missed even at autopsy because macroscopic changes may be slight, and the infiltration may be confined to the spinal meninges. The lack of localizing signs and the protean nature of the disease often leads the uninitiated clinician astray, but paradoxically it is the diversity of symptoms and multifocal findings that should alert one to the diagnosis. Patients often offer myriad complaints, sometimes minor or vague, and frequently they are only elicited by
MENINGEAL CARCINOMATOSIS
TABLE Vi
Meningeal
IN SMALL CELL CARCINOMA
OF LUNG-ARONEY
ET AL.
Relapses-Results Responders No.
Results of therapy lntrathecal methotrexate alone Plus cranial radiotherapy (R,T) Plus cranial R,T and cytosine arabinoside No therapy given Duration of response (wk) Median Duration of survival from time meningial carcinomatosis diagnosed (wk) Over-all Responders Nonresponders
direct questioning [ 161. The meninges and brain as pharmacologic sanctuary sites have recently become increasingly important in the management of several malignancies, particularly the leukemias and lymphomas. Small cell carcinoma of the lung is now also of particular relevance in this context as intensive chemotherapy has improved the prognosis of this subtype. The high response rates and prolonged survival patterns of our patients are comparable with the better combinations reported by other workers, but the extreme prominence of relapses within the central nervous system, especially meningeal, highlights the sanctuary site problem. Meningeal involvement in small cell carcinoma of the lung must now be considered a sanctuary site of equal importance to cerebral metastases, particularly as the outlook for patients from the time of relapse within the central nervous system is not good, and is particularly poor for those with meningeai carcinomatosis. Although gratifying and worthwhile control of symptoms and neuroiogic deficits was achieved in 50 percent of our patients with meningeal carcinomatosis, the longest response was less than two months and the best survival was only a little over four months (Table VI). Prevention would obviously be preferable. Our patients received neither prophylactic cranial or spinal irradiation, nor prophylactic intrathecai therapy. Furthermore, none of the chemotherapeutic agents in the doses used are known to achieve therapeutic concentrations in the central nervous system. Certainly, prophylactic cranial irradiation appears to reduce the incidence of cerebral mass lesion relapses, but it confers no definite survival advantage [6,7] and has not yet been evaluated for its possible influence on the development of spinal meningeal disease. Chemotherapeutic agents which readily cross the blood-brain barrier, such as l-(2-chloroethyi)-3-cyclohexyl-l-ni-
Range
No. 3 1 1
5 4 1 1 a
4-10
5 11 2
2-10 8-17 4-6
%60 25 100
trosourea (CCNU) and procarbazine have not been demonstrated to influence the incidence of central nervous system relapses [ 171, however, the ultimate role of these agents and of high dose MTX programs will require randomized prospective trials. Prophylactic spinal irradiation and/or intrathecal cytotoxic therapy remain essentially unevaluated, understandably because of the problems of myelosuppression and other toxicities particularly with concurrent systemic chemotherapy. It is not possible at this stage to determine the exact basis for the inordinately high incidence of meningeal carcinomatosis reported here. it would seem likely that more than one factor is operative. Firstly, the extended survival resulting from high levels of systemic disease control, allowing time for the progression of meningeal disease to the point of clinical detection. Even autopsy evaluations of therapy programs, less effective in extending survival than is now the rule, have repeatedly demonstrated central nervous system involvement in from 28 to 55 percent of the patients. Secondly is the absence of systemic chemotherapeutic agents with significant central nervous system penetration. Thirdly is the possibility that the cytotoxic agents in combination might actively predispose to the development of meningeai carcinomatosis by some as yet unknown mechanism. Finally, it is possible, that the incidence has been underreported to date. Certainly, attention to all possible mechanisms of meningeal carcinomatosis relapse will be required if effective prophylactic programs are to be designed. There is also a pressing need for more effective management of meningeal carcinomatosis relapse. Combination intrathecal cytotoxic therapy warrants further evaluation, but major advances will probably require the availability of new effective agents and/or better tolerated, yet more effective, radiotherapeutic approaches for control of leptomeningeal disease.
July 1991
The American Journal of Medicine
Volume 71
31
MENINGEAL CARCINOMATOSIS
1N SMALL CELL CARCINOMA OF LUNG-ARONEY
ET AL.
REFERENCES 1.
32
Roswit B, Patno ME, Rapp R. et al.: The survival of patients with inoperable lung cancer. A large-scale randomized study of radiotherapy versus placebo. Radiology 1966; 90: 686-697. Greco FA, Einhorn LH, Richardson RL, Okfham RK: Small cell lung cancer: progress and perspectives. Semin Oncol 1976; 5: 323-335. Einhorn LH, Bond WH, Hornback N, Beng-Tek J: Long-term results in combined-modality treatment of small cell carcinoma of the lung. Semin Oncol 1976; 5: 309-313. Bunn PA, Nugent JL, Matthews MJ: Central nervous system metastases in small cell bronchogenic carcinoma. Semin Oncol 1976; 5: 314-322. Nugent JL, Bunn PA, lhde DC, et al.: Diagnosis and treatment of CNS complications of small cell bronchogenic carcinoma (SCBC). Proceedings of the American Association for Cancer Research and the American Society of Clinical Oncology 1978; 19: 116. Jackson DV, Richards F, Cooper MR. et al: Prophylactic cranial irradiation in small cell carcinoma of the lung: a randomized study. JAMA 1977; 237: 2730-2733. Holoye PY, Sam&s ML. Lanzotti VJ, Smith T, Barkley HT Jr: Combination chemotherapy and radiation therapy for small cell carcinoma. JAMA 1977; 237: 1221-1224. Chan HC, Carey RW: Small cell anaplastic carcinoma of lung. Reappraisal of current management. Cancer 1976; 37: 2651-2657. Hornback NB, Einhorn L, Skidnia H, Joe BT, Krause M, Furnas B: Oat cell carcinoma of the lung. Early treatment results
July 1961
The American Journal of Medicine
Volume 71
10.
11.
12.
13. 14.
15.
16.
17.
of combination radiation therapy and chemotherapy. Cancer 1976; 37: 2658-2664. Matthews MJ, Gazdar AF, lhde DC, et al.: Histologic subtypes of small cell carcinoma of the lung and their clinical significance. Proceedings of the American Society of Clinical Oncology 1978; 19: 397. Brereton HC, O’Donnell JF, Kent CH. et al.: Spinal meningeal carcinomatosis in small-cell carcinoma of the lung. Ann Intern Med 1978; 88: 517-519. Elliott KAC, et al.: Physiological salt solutions for brain surgery. Studies of local pH and pial vessel reactions to buffered and unbuffered isotonic solutions. J Neurosurg 1949; 6: 140-152. Lewis RC, et al.: Clinical uses of an artificial cerebrospinal fluid. J Neurosurg 1950; 7: 256-260. Eberth CJ: Zur Entwickluna des Eoithetioms (Cholesteatoms) der Pia und der Lunge. Vi&how Arch ‘1870; 49: 51-63. Lilienfeld J, Benda C: Fall von Metastischer Karcinose der Neivenund Hirnhaute. Berlin Klin Wochenschr 1901; 38: 729-730. Olsen ME, Chernlk NL, Posner JB: Infiltration of the leptomeninges by systemic cancer. Arch Neurol 1974; 30: 122-137. Nugent JL, Bunn PA, lhde DC, et al.: Diagnosis and treatment of CNS complications of small cell bronchogenic carcinoma. Proceedings of the American Association for Cancer Research and the American Society of Clinical Oncology 1978; 19: 116.