Printer Friendly

Dural Metastases: A Retrospective Surgical and Autopsy Series.

A Retrospective Surgical and Autopsy Series

Dural metastatic disease is one of the least frequent and least studied of all the patterns of cancer spread. Only a few reports on metastatic dural disease exist in the literature, and these are virtually exclusively autopsy studies from the 1970s or earlier.[1-6] Most of these articles document the incidence of metastatic tumor spread to all craniospinal compartments, including brain parenchyma, spinal cord, leptomeninges, and dura. Dural metastatic disease has seldom been the primary focus of investigation.[7]

The most quoted article is that of Posner and Chernik,[1] who studied intracranial metastases in 2375 patients with systemic cancer who underwent autopsy between 1970 and 1976. Metastases to the epidural or subdural surfaces of the cranial vault, collectively considered "dural metastases," were identified in 9% of their cases. In 4% of these patients, metastases to the dura were isolated and represented the only intracranial tumor involvement. Dural metastatic disease was approximately equal in frequency to that for another rare pattern of central nervous system cancer spread, leptomeningeal carcinomatosis. Primary tumor types in patients with isolated dural disease included breast, prostate, neuroblastoma, Hodgkin disease, non-Hodgkin lymphoma, lung, and melanoma, in decreasing order of frequency.

Earlier studies by Lesse and Netsky[2] in 1954 and Meyer and Reah[3] in 1953 had reported a different spectrum of offending malignant neoplasms than that reported by Posner and Chernik[1] Lesse and Netsky[2] emphasized prostatic carcinoma, multiple myeloma, malignant lymphoma, and leukemia. Like Posner and Chernik, Meyer and Reah[3] found that dural metastases constituted 9% of their total central nervous system metastases, but the tumor types differed, with carcinomas of the lung, prostate, stomach, and breast equally represented. Unlike other series, Meyer and Reah reported that 20% of their cases with diffuse dural metastases had coincident leptomeningeal carcinomatosis and nearly half had associated subdural hemorrhages.

Later studies showed additional differences, with an incidence of dural metastases as low as 3%4 or an inordinately high percentage of dural metastases due to breast cancer (51%),[5] breast and lung cancers,[6] or lymphomas and sarcomas.[7]

Other studies have been case reports or have detailed small numbers of patients with specific types of cancers metastatic to the dura, such as breast[8] or prostate[9-13] carcinoma, Ewing sarcoma,[14] medulloblastoma,[15] or leukemia.[16] Still others highlight special features such as the coexistence of subdural metastatic tumors with subdural hematomas.[17-19]

Surgical series of patients with dural metastatic disease are even less frequent than autopsy series, with a single, recent report by Rumana et al.[20] In their 1998 article detailing their experience with 22 patients who underwent surgery for solitary metastatic brain tumors with dural extension, the most common tumor types were breast, lung, and renal cell carcinomas. They noted a median patient survival time of 11 months following resection of the dural tumor, a survival equal to that of the comparison group of surgical patients with metastatic brain tumors without dural extension (10 months). Their study suggested that patients with metastatic brain tumors with dural extension did not necessarily have more end-stage disease than those patients with brain parenchymal metastases alone and could be suitable surgical candidates under appropriate clinical circumstances.

I sought in this study to document my experience with dural metastatic disease and to compare it with the older autopsy series[1-7] and the surgical series of Rumana et al.[20] My impression at the beginning of this study was that I had seen a large number of unusual tumor types metastatic to dura, other than breast and prostate carcinomas. Unlike these previous studies, I included both autopsy and surgical cases in the same report for comparison purposes, although the focus of the study was necessarily somewhat different for the 2 patient groups.

MATERIALS AND METHODS

The study included all consultation and in-house autopsy and surgical cases in the files with epidural, intradural, and/or subdural metastatic tumor deposits clearly documented as to site of tumor and confirmed on examination of the histologic sections. Excluded from this study were all cases with direction extension of tumor to dura from tumors primary to the intracranial compartment, such as glioblastomas, oligodendrogliomas, or pituitary adenomas, and direct extension of aggressive head and neck cancers to the dura. Hematologic neoplasms (leukemias, lymphomas, intravascular lymphomatosis) were included, paralleling older series.

The period of study for autopsy cases was 1982 through 1999 at University Hospital, University of Colorado Health Sciences Center, Denver. These records included autopsy consultation cases from the Denver Veteran's Administration Hospital, which services a large, older male patient base, with significant numbers of patients with prostatic carcinoma. Although neonatal autopsies were included in these autopsied cases, pediatric patients were underrepresented in this autopsy series, since such autopsies are Conducted at the Children's Hospital. Because of the considerably lower rate of full spinal cord removal than cranial vault examination in all of the autopsy cases, the autopsy series included few cases with well-documented subdural or epidural disease at spinal cord levels.

The period of study for surgical cases was 1990 to August 1, 2000. Accrual of cases was accomplished via both manual review and a computer-based word search of our surgical pathology records, using the key words brain, metastatic, and dura to retrieve potential cases from the computer database. Cases with large, virtually exclusive bony involvement, especially of the vertebral bodies, and only scant, focal extension to the epidural surface were not considered "dural involvement" for the purposes of this study.

The focuses of the study for autopsy cases were on identifying the histologic types of cancers, the epidural versus subdural location of metastatic disease, tumor size, and diffuse versus nodular pattern of dural tumor spread. Coassociation of dural disease with the presence of extradural metastatic disease in systemic bones, skull bone, brain parenchyma, leptomeninges, and lung was recorded. Whether dural disease was likely due to direct extension from contiguous skull metastases was determined from the autopsy records and the finding of epidural tumor deposits directly overlying the subdural tumor.

The focuses of the study for surgical cases were on identifying histologic types of cancers, presence or absence of contiguous disease in brain or skull, and postoperative patient survival times. Whether there were additional systemic tumor metastases to noncalvarial bones and lung usually could not be determined from the records. Follow-up information was obtained from our Medical Records Department; survival was recorded as the last reported time the patient was known to be living. We were unable to determine the extent of systemic tumor burden at death or cause of death, since most patients in the surgical series were neither treated at our hospital during the terminal stages of their illnesses nor autopsied at our facility.

RESULTS

The details of the autopsy case series results are given in Table 1. Twelve of the 27 cases were from expected histologic tumor types, including 5 cases of breast carcinoma (Figure 1, A and B) and 7 of prostate carcinoma. Lung cancer was not overly represented (3 cases; Figure 1, C). Several lymphomas also were part of the group, paralleling some older series. However, unusual tumor types metastatic to dura occurred throughout the study. These included 3 cases of cervical carcinoma (Figure 1, D), the most recent one from a patient with acquired immuno-deficiency syndrome, and single cases of laryngeal squamous cell carcinoma, adenosquamous carcinoma of the bladder, intravascular lymphomatosis, Ewing sarcoma, and ocular melanoma (Figure 2, A through C).

[ILLUSTRATIONS OMITTED]
Table 1. Dural Metastases: Autopsy Case Series(*)

 Extradural
 Metastatic
 Disease

Case No./ Type of Location of Systemic Skull
Sex/Age, y Neoplasm Dural Metastases Bones Bone

1/F/32 Cervical Large subdural Yes Yes
 carcinoma plaquelike tumor
 masses with small
 overlying epidural
 tumor
2/F/45 Chronic Diffuse cranial Yes Yes
 myelogenous studding
 leukemia
3/M/55 Lung Single 1.5-cm Yes Yes
 carcinoma, epidural tumor mass
 adeno-
 carcinoma
4/F/59 Breast Diffuse Yes Yes
 carcinoma supratentorial and
 infratentorial
 nodular tumor
 studding
5/M/71 Prostate Diffuse 0.4- to Yes Yes
 carcinoma 2.0-cm subdural
 tumor studding plus
 subdural empyema
6/M/69 Laryngeal Single large globular N/A N/A
 squamous subdural tumor,
 cell small nearby
 carcinoma plaquelike
 meningioma
7/M/69 Prostate Diffuse subdural Yes Yes
 carcinoma 0.2- to 0.8-cm
 tumor studding
8/F/55 Cervical Dural infiltration by Yes No
 carcinoma tumor
9/F/56 Breast Single parasagittal Yes N/A
 carcinoma 3.5 x 3.5-cm
 subdural tumor;
 nearby
 parasagittal 0.3 x
 0.4-cm meningioma
10/F/52 Breast Diffuse subdural Yes N/A
 carcinoma tumor studding
11/M/66 Prostate Single left parietal Yes Yes
 carcinoma subdural 5 x 4 x
 3 cm tumor with
 adjacent chronic
 and subacute
 subdural hematoma
12/M/47 Lung Diffuse epidural, N/A N/A
 carcinoma, intradural, and
 small cell subdural tumors
 (large cell with scant subdural
 variant) hemorrhage
13/M/76 Lung Single 2-cm left Yes N/A
 carcinoma, subdural occipital
 large cell tumor nodule
 undiffe-
 rentiated
14/M/61 Large cell Diffuse dural Yes Yes
 lymphoma involvement
15/M/50 Prostate Subdural 4.0 x 5.0-cm Yes Yes
 carcinoma right skull base
 and 1.0 x 1.0-cm
 convexity left
 temporal subdural
 tumors, with right
 thin chronic
 subdural hematoma
16/F/53 Breast Single 7-cm left Yes Yes
 carcinoma parietal bony tumor
 with dural and
 subdural extension
17/M/63 Prostate Dural tumor nodule Yes Yes
 carcinoma in region of sella/
 pituitary gland
18/M/67 Prostate Diffuse subdural 0.1- Yes Yes
 carcinoma to 0.3-cm tumor
 studding, small
 meningioma
19/M/66 Adenosquamous Multiple right No No
 carcinoma subdural occipital
 of tumor nodules
 gallbladder
20/F/64 Chronic Multiple cranial and Yes N/A
 myelomono- spinal cord
 cytic subdural chloromas,
 leukemia with massive
 falcine chloroma
21/F/39 Breast Multiple subdural (3) Yes No
 carcinoma and epidural (3)
 0.5-cm tumors
22/M/66 Multiple Subdural cervical Yes Yes
 myeloma cord and epidural
 thoracic level
 tumor, bilateral
 cranial chronic
 subdural
 neomembranes
23/F/21 Cervical Single falcine 9.0 x Yes N/A
 carcinoma 0.3 x 1.5-cm
 subdural metastasis
 and microscopic
 epidural spinal
 cord dural tumor
24/M/26 Ewing Diffuse, massive Yes Yes
 sarcoma epidural and
 subdural plaquelike
 tumor nodules
25/M/47 Prostate Bifrontal subdural Yes Yes
 carcinoma 7.0-cm tumor and
 diffuse 0.5-cm
 subdural tumor
 studding
26/M/62 Intravascular Subdural right Yes N/A
 lymphoma- occipital tumor and
 tosis, with acute subdural
 cerebral hematoma 2 x 3 cm
 micro-
 infarctions
 and macro-
 infarctions
27/M/75 Ocular Single residual left No No
 melanoma, occipital 2 x 2 x
 left 1-cm subdural tumor
 nodule (previous
 surgical removal of
 tuberculum sella
 and falcine
 subdural metastases)

 Dural
 Disease
 Extradural Metastatic Likely
 Disease Due to
 Direct
 Brain Extension
Case No./ Parenchymal Leptomeningeal Lung From Skull
Sex/Age, y Metastases Carcinomatosis Involvement Metastases

1/F/32 Yes (direct No Yes; lymph- Yes
 extension angitic
 into tumor
 right involvement
 parietal
 lobe)
2/F/45 No No No Yes
3/M/55 Yes (single No Yes; primary Yes
 0.4-cm tumor and
 frontal left lung
 lobe) metastasis
4/F/59 No No No Possible
 (epidural
 tumor
 identi-
 fied)
5/M/71 No No No ...
6/M/69 N/A N/A Yes; left ...
 upper lobe
 metastasis
7/M/69 No No Yes; lymph- ...
 angitic and
 pleural
 tumor
 involvement
8/F/55 N/A N/A No
9/F/56 Yes (micro- No Yes; paren- ...
 scopic chymal and
 disease pleural
 near tumor
 dural involvement
 meta-
 stasis)
10/F/52 Yes (micro- No Yes; lymph- ...
 scopic angitic and
 meta- intra-
 stases) vascular
 tumor
 involvement
11/M/66 N/A N/A Yes No
 (epidural
 surface
 free of
 tumor)
12/M/47 N/A N/A Yes; primary
 tumor
13/M/76 Yes (single No Yes; primary
 right tumor and
 occipital left lung
 lobe me- metastases
 tastasis)
14/M/61 N/A N/A Yes ...
15/M/50 Yes (micro- No Yes; paren- Yes
 scopic chymal and
 tumor in- pleural
 volvement tumor
 adjacent involvement
 to
 subdural
 meta-
 stasis)
16/F/53 Yes (micro- No Yes; lymph- Yes
 scopic angitic and
 tumor in- intra-
 volvement vascular
 adjacent tumor
 to involvement
 subdural
 meta-
 stasis)
17/M/63 Yes (micro- No Yes; lymph- Yes
 scopic angitic and
 intra- intra-
 vascular vascular
 tumor tumor
 emboli) involvement
18/M/67 No No Yes; lymph- ...
 angitic
 tumor
 involvement
19/M/66 Yes (4 No Yes; lymph- ...
 macro- angitic and
 scopic intra-
 meta- vascular
 stases) tumor
 involvement
 with
 pulmonary
 micro-
 infarctions
20/F/64 Yes (intra- No Yes; ...
 vascular extensive
 tumor leukemic
 only) infil-
 tration of
 alveoli and
 inter-
 stitium
21/F/39 Yes (micro- No Yes; lymph- ...
 scopic) angitic and
 pleural
 tumor
 involvement
22/M/66 No No No ...
23/F/21 Yes No Yes; paren-
 (multiple chymal and
 macro- pleural
 scopic tumor
 meta- involvement
 stases)
24/M/26 No No Yes; paren- Yes
 chymal and
 pleural
 tumor
 involvement
25/M/47 No No Yes; lymph- Yes (bi-
 angitic, frontal
 intra- mass)
 vascular,
 and pleural
 tumor
 involvement
26/M/62 Yes Yes (at least Yes; intra- No
 focally vascular
 with and paren-
 associated chymal
 subarachnoid tumor
 hemorrhage) involvement
27/M/75 No Focal lepto- Yes; paren- No
 meningeal chymal
 spread tumor
 around involvement
 optic nerve
 only

(*) N/A indicates not available.


Most patients for whom data could be obtained had systemic bony disease, and most also had calvarial metastases. The exceptions were the patients with ocular melanoma and carcinoma of the gallbladder. As best as could be determined, however, those with calvarial metastases did not show their dural disease immediately subjacent to grossly apparent bony tumor involvement. Since the skull was virtually never extensively sectioned and histologically examined, we could not rule out microscopic disease in the adjacent skull as a cause of contiguous dural disease. Pulmonary involvement, especially lymphangitic spread, was almost as frequent a finding as bony disease in patients with dural metastatic disease. Twenty-two of 27 cases showed neoplastic lung involvement at autopsy, of which nearly half of these were recorded as having intravascular or lymphangitic pulmonary tumor. Coexistent brain parenchymal metastases were found in less than half of cases (13 of 27); large macroscopic brain metastases (seen in cases 1, 3, 13, 19, 23, 26) were less frequently identified than were microscopic brain metastases.

Diffuse studding of dura was seen primarily with breast (Figure 1, A) and prostatic carcinomas, although several patients with these tumors had large single metastases (Table 1). Although the case numbers were too few for strong conclusions, our hematologic malignant neoplasms and a single case of sarcoma also showed diffuse disease. Several patients showed both epidural and subdural metastatic spread (Table 1). Usually the subdural component was more conspicuous, although in the patients with Ewing sarcoma and small cell lung carcinoma (Figure 1, C) nearly equally sized subdural and epidural tumor deposits were found. The subtlety of the epidural component in some cases suggested that it might be easily overlooked (Figure 1, A and B).

Five of 27 patients had coexistent subdural hemorrhages (autopsy patients 11, 12, 15, 22, and 26), and autopsy patient 12 also had hemorrhagic epidural metastases (Figure 1, C). In several of these cases, the blood was minimal and would be highly unlikely to be of clinical relevance. Coexistent meningiomas were found in 3 patients (autopsy patients 6, 9, and 18), but only in autopsy patient 6 had there been difficulty at the time of brain cutting in distinguishing grossly which mass was the metastasis and which was the meningioma. One patient appeared on both the surgical and autopsy series (patient 27 in autopsy series is cases 20 and 21 in the surgical series). Several cases of breast, prostate, and cervical carcinomas had large subdural deposits of tumor, with microscopic (autopsy cases 9, 10, 15, and 16) or macroscopic (autopsy case 1) extension into the immediately underlying brain parenchyma.

Our surgical cases contained an even greater spectrum of tumor types than did our autopsy series (Table 2). Several patients had 2 different dural metastases removed several months apart (surgical cases 1 and 2 and cases 20 and 21 are from the same patient). Surgical patient 33 just recently returned to our hospital for removal of another dural metastasis 2 years after her previous procedure for removal of dural-based metastases. In both instances, the dural tumors occurred directly in the trajectory of the original biopsy site of her pineal tumor, according to her neurosurgeon. No other dural disease in other sites was found radiographically or intraoperatively, except near the site of her previous biopsy. Pathologically, no tumor deposits were found within dural venous sinuses since they often are involved subsequent to bony metastases.[8,9] Hence, these dural-based metastases were clinically and pathologically consistent with postsurgical implantation metastases.
Table 2. Dural Metastases: Surgical Case Series

 Contiguous
 Location Malignant
Case of Disease:
No./Sex/ Type of Dural Brain or Postoperative
Age, y Neoplasm Metastases Skull Survival, wk

1/M/36 Melanoma Cranial(*) Yes, brain 29
2/M/36 Melanoma Cranial(*) Yes, brain 12
3/F/41 Breast Cranial Yes, brain 39
 carcinoma
4/F/26 Breast Cranial Yes, skull 338
 carcinoma
5/M/42 Colon Cranial Yes, skull 12
 carcinoma
6/F/67 Endometrial Spinal, L1 No 43
 carcinoma level
7/M/34 Melanoma Cranial Yes, brain 14
8/F/49 Melanoma Cranial, Yes, brain 6
 with
 subacute
 subdural
 hematoma
9/M/59 Melanoma Cranial, Yes, brain 16
 with
 acute
 subdural
 hematoma
10/F/51 Breast Cranial Yes, skull 20
 carcinoma
11/M/65 Prostate Cranial Yes, skull 8
 carcinoma
12/M/52 Lung Cranial Yes, brain 71
 carcinoma
13/F/29 Medull- Cranial No 5
 oblastoma
14/M/40 Renal cell Cranial No [is greater than
 carcinoma or equal to]
 215; Alive as of
 November 1999
15/F/50 Breast Cranial Yes, skull [is greater than
 carcinoma or equal to]
 156; Alive as of
 April 1999
16/M/64 Renal cell Spinal No [is greater than
 carcinoma (epi- or equal to]
 dural), 199; Alive as of
 thoracic July 2000
17/M/64 Adeno- Cranial Yes, brain 21
 carcinoma,
 unknown
 primary
18/1:/62 Adeno- Cranial Unclear 14
 carcinoma,
 unknown
 primary
19/F/34 Pineal pa- Cranial(*) No [is greater than
 renchymal or equal to]
 tumor with 120; Alive as of
 inter- June 2000
 mediate
 diffe-
 rentiation
20/M/75 Ocular Cranial, No [is greater than
 melanoma falx(*) or equal to]
 24
21/M/75 Ocular Cranial, No [is greater than
 melanoma tuber- or equal to]
 culum 22
 sellae(*)
22/F/36 Breast Cranial No 34
 carcinoma
23/M/47 Carcinoma of Cranial, No 5
 stomach with
 acute
 subdural
 hematoma
24/M/45 Large cell Cranial No 48
 lymphoma
25/F/66 Large cell Not No 9
 lymphoma specified
26/F/53 Lung Cranial Yes, brain [is greater than
 carcinoma or equal to]
 38; Alive as of
 June 2000
27/F/48 Intra- Cranial No Lost to follow-up
 vascular
 lymphoma-
 tosis
28/F/70 Breast Spinal, No [is greater than
 Carcinoma T5-6 or equal to]
 20; Alive as of
 April 2000
29/F/62 Lung Cranial No 4
 carcinoma
30/F/36 Cervical Cranial No [is greater than
 carcinoma or equal to]
 12; Alive as of
 August 1, 2000
31/M/36 Malignant Cranial Yes, skull [is greater than
 melanoma or equal to]
 19; Alive as of
 August 1, 2000
32/M/19 Germinoma, Cranial No, but [is greater than
 pineal possible or equal to]
 origin pineal 5; Alive as of
 re- August 1, 2000
 currence
33/F/36 Pineal pa- Cranial(*) No, but [is greater than
 renchymal nearby or equal to]
 tumor with brain 2; Alive as of
 inter- lesion August 1, 2000
 mediate
 diffe-
 rentiation

(*) Cases 1 and 2 and cases 20 and 21 are the same patients; each
patient had 2 surgical procedures several months apart for removal of
dural-based metastases; cases 19 and 33 are the same patient who had 2
surgical procedures 2 years apart for removal of dural-based
metastases.


Three cases in the surgical series were also associated with acute or subacute subdural hematomas of varying clinical relevance (surgical cases 8, 9, and 23). Subdural hematoma prompted surgical intervention in surgical case 23, and the finding of cancer was unexpected. In the surgical case of ocular melanoma, the preoperative diagnosis for the dural-based metastases had been meningioma based on the premortem radiographic studies (Figure 2, A and B).

Contiguous malignant disease in underlying brain was particularly frequent with nonocular melanoma, whereas contiguous skull, but not contiguous brain, metastatic disease was found in most patients with breast carcinoma. Prostate carcinoma was conspicuously infrequent in the surgical case series, compared with the autopsy series, with only one case. Again, several unusual tumor types were found, including endometrial, stomach, and cervical carcinomas, an ocular melanoma, and germinoma metastatic from pineal gland.

Postoperative survival was extremely varied, and although overall was generally poor, notable exceptions were found. Tumor survival varied considerably according to histologic tumor type and probably to the presence and extent of systemic tumor involvement, making calculation of an overall survival rate for dural metastatic disease of limited usefulness. The worst survival was for patients with cutaneous, nonocular melanoma. Several patients with breast cancer and the 2 patients with renal cell carcinoma have enjoyed exceptionally long survival times following removal of their dural metastases (Table 2).

COMMENT

Several conclusions can be drawn from this large autopsy series of patients with metastatic dural disease, particularly regarding mode of tumor spread to dura, other types of coexistent metastatic disease in these patients, and types of tumors comprising the group. Mode of tumor spread to the dura appeared to vary according to histologic type of the primary tumor. Direct tumor extension to the dura from calvarial metastases was seen with lung, cervical, some prostatic, and breast carcinomas and Ewing sarcoma. In breast carcinomas such bone-to-dura spread has been hypothesized by Tsukada et al[8] to arise from dissemination of tumor cells from bony metastases, especially the vertebral bodies, with retrograde reflux of tumor cells into veins and venous plexi and subsequently into the dura.

Hematogenous spread of cancer cells to dura from some systemic malignant neoplasms appeared likely. Most cases of dural metastatic disease in the autopsy series were associated with lung involvement, especially lymphangitic and intravascular pulmonary tumor spread, suggesting heavy vascular tumor burden in these patients. Hematogenous tumor spread to dura was also likely in the autopsy case of gallbladder carcinoma, since there was no associated bony metastatic disease and the dural venous sinuses overlying the dural metastasis were free of metastatic tumor cells. In contrast, tumor cells are often found in dural venous sinuses in patients with the bone-to-dura route of tumor spread.[9]

In several of these cases there appeared to be spread of a dural metastasis to underlying brain parenchyma, a pattern not emphasized in previous studies. In several autopsy cases with breast and prostatic cancers metastatic to dura, there was an adjacent small microscopic extension of tumor into the superficial cortex. Overall, however, macroscopic parenchymal brain metastases associated with dural tumor deposits were seen in a minority of cases (6 of 27) in my series and have been observed by others to be particularly uncommon with prostatic carcinoma.[12,13]

Finally, some cases seemed to display the spread of brain parenchymal metastasis to the dura. Secondary attachment to dura from superficially located brain parenchymal metastases likely occurred in most cases of nonocular malignant melanoma.

Types of tumors metastatic to dura differed somewhat between the autopsy and surgical series, and the study of both patient groups provided the opportunity to compare and contrast dural metastatic disease in these 2 populations. Most notable was the significant presence of prostatic disease metastatic to dura in the autopsy series (7 of 27 cases) versus its near absence in the surgical series (1 of 33 cases). The high number of prostatic carcinomas in the autopsy series likely reflects the autopsy cases being partially drawn from a hospital with a large adult male patient load (ie, Veteran's Administration Hospital). Also important, however, may be the fact that metastatic dural disease in prostate carcinoma is indeed fairly characteristic of late, advanced clinical stages of this type of cancer, and such patients do not make good surgical candidates because of their projected short survival times or poor clinical statuses.

Nonocular malignant melanoma cases appeared only in the surgical series and only in the early years of this study. This reflects the clinical practice at our institution in those early years to treat patients with malignant melanoma and brain metastases with aggressive surgery. Some of these melanoma patients underwent multiple craniotomies for removal of metastatic tumors, such as surgical cases 1 and 2 (Table 2). These patients with nonocular malignant metastatic melanoma also had the shortest postoperative survival times following removal of their dural metastases. Since that time, the use of aggressive surgical removal of brain and dural melanoma metastases has been discontinued, and small melanoma metastases are now preferentially treated with stereotactic radiosurgery (K. Lillehei, MD, oral communication, 2000).

Unlike the older autopsy series in the literature of Posner and Chernik,[1] and especially Takakura et al[5] and France,[6] my series did not contain an overwhelming predominance of breast cancers as the cause of dural metastases. This may reflect differences in treatment, survival times, or tendency to forgo autopsy examination for patients with breast cancer in the 1980s and 1990s versus the earlier years studied by these other authors. My autopsy series also contained a greater percentage of unusual malignant neoplasms. A few of the rare tumor types seen in my series were, in fact, identified in the autopsy series of Posner and Chernik,[1] but were overshadowed by the numbers of patients with breast and prostate cancer and dural metastases.

Several patients with cervical carcinoma metastatic to dura in my series were of particular interest. Patient 30 in the surgical series was identified as having dural metastatic disease when her case was presented at our institutional brain tumor board, following discussion that this pattern of spread was possible for cervical cancer.

Only a few of the patients from this surgical or autopsy series displayed coexistent subdural hematomas or meningiomas, and only 2 had coincident leptomeningeal carcinomatosis. These numbers did not approach the high figures quoted by Meyer and Reah.[3] Cases of dural metastatic disease with these additional features have been the subject of reports in the literature.[17-19]

Several of the patients in the surgical series experienced extended postoperative survival times of greater than 2 years, particularly 2 individuals with renal cell carcinoma and 2 with breast carcinoma (Table 2). This corroborates the impression of Rumana et al[20] that some patients with these tumor types may achieve good postoperative survival times and surgical removal is clinically advisable. The patient with surgical implantation metastases from a pineal parenchymal with intermediate differentiation has also survived 2 years from her first dural metastasis. A conclusion that can be drawn from this finding of favorable survival in some surgical patients is that the presence of dural disease does not necessarily reflect end-stage, advanced, widespread systemic cancer involvement. Patients whose dural metastases arise from routes other than hematogenous spread or spread from bony metastases may be particularly able to achieve longer Survival times, if surgically treated.

The author gratefully acknowledges Kevin Lillehei, MD, Robert Breeze, MD, and Stephen J. Pfister, MD, who furnished additional information on several patients in this series; Mrs Virginia McCullough for manuscript preparation; and Mr Bob McCullough for photographic expertise.

References

[1.] Posner JB, Chernik NL. Intracranial metastases from systemic cancer. In: Schoenberg BS, ed. Advances in Neurology. New York, NY: Raven Press; 1978; 19:579-92.

[2.] Lesse S, Netsky MG. Metastasis of neoplasms to the central nervous system and meninges. Arch Neurol Psychiatry. 1954;72:133-153.

[3.] Meyer PC, Reah TG. Secondary neoplasms of the central nervous system and meninges. Br J Cancer. 1953;7:438-448.

[4.] Willis RA. The Spread of Tumours in the Human Body. London, England: Butterworth & Co; 1973.

[5.] Takakura K, Sano K, Hojo S, Hirano A. Metastatic Tumors of the Central Nervous System. Tokyo, Japan: Igaku-Shoin Ltd; 1982.

[6.] France LH. Contribution to the study of 150 cases of cerebral metastases, II: neuropathological study. J Neurosurg Sci. 1975;19:189-210.

[7.] Sieben G, deReuck J, deCoster W, vanderEecken H. Cranial subdural metastases: a clinicopathological study. Clin Neurol Neurosurg. 1981;83:81-86.

[8.] Tsukada Y, Fouad A, Pickren JW, Lane WW. Central nervous system metastasis from breast carcinoma. Cancer. 1983;52:2349-2354.

[9.] Pfister S, Kleinschmidt-DeMasters BK. Dural metastases from prostatic adenocarcinoma. J Urol Pathol. 1995;3:119-127.

[10.] Penley MW, Kim YC, Pribram HFW. Subdural metastases from prostatic adenocarcinoma. Surg Neurol. 1981 ;16:131-134.

[11.] Lippman SM, Buzaid AC, Iacono RP, et al. Cranial metastases from prostate cancer simulating meningioma: report of two cases and review of the literature. Neurosurgery. 1986;19:820-823.

[12.] Catane R, Kaufman J, West C, Merrin C, Tsukada Y, Murphy GP. Brain metastasis from prostatic carcinoma. Cancer. 1976;38:2583-2587.

[13.] Castaldo JE, Bernat JL, Meier FA, Schned AR. Intracranial metastases due to prostatic carcinoma. Cancer. 1983;52:1739-1747.

[14.] Shuper A, Cohen IJ, Mor C, Ash S, Kornreich L, Zaizov R. Metastatic brain involvement in Ewing family of tumors in children. Neurology. 1998;51:1336-1338.

[15.] Koenig GH. Subdural spread of medulloblastoma: case report. J Neurol Neurosurg Psychiatry. 1971;34:436-438.

[16.] Azzarelli B, Roessmann U. Pathogenesis of central nervous system infiltration in acute leukemia. Arch Pathol Lab Med. 1977;101:203-205.

[17.] Ambiavagar P-C, Sher J. Subdural hematoma secondary to metastatic neoplasm. Cancer. 1978;42:2015-2018.

[18.] Leech RW, Welch FT, Ojemann GA. Subdural hematoma secondary to metastatic dural carcinomatosis. J Neurosurg. 1974;41:610-613.

[19.] McKenzie CR, Rengachary SS, McGregor DH, Dixon AY, Suskind DL. Subdural hematoma associated with metastatic neoplasms. Neurosurgery. 1990;27: 619-625.

[20.] Rumana CS, Hess KR, Shi WM, Sawaya R. Metastatic brain tumors with dural extension. J Neurosurg. 1998;89:552-558.

Accepted for publication January 23, 2001.

From the Departments of Pathology and Neurology, The University of Colorado Health Sciences Center, Denver, Colo.

Presented in part in abstract form at the 76th annual meeting of the American Association of Neuropathologists, Atlanta, Ga, June 2000.

Reprints: B. K. Kleinschmidt-DeMasters, MD, Department of Pathology B216, University of Colorado Health Sciences Center, 4200 E Ninth Ave, Denver, CO 80262 (e-mail: BK.DeMasters@UCHSC.edu).
COPYRIGHT 2001 College of American Pathologists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

 Reader Opinion

Title:

Comment:



 

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:study of cancer spread pattern
Author:Kleinschmidt-DeMasters, B. K.
Publication:Archives of Pathology & Laboratory Medicine
Geographic Code:1USA
Date:Jul 1, 2001
Words:5067
Previous Article:Computerized Image Analysis of p53 and Proliferating Cell Nuclear Antigen Expression in Benign, Hyperplastic, and Malignant Endometrium.
Next Article:Evaluation of Characteristics Associated With Acute Splenitis (Septic Spleen) as Markers of Systemic Infection.
Topics:


Related Articles
Understanding Cancer's Spread.
Cytokeratin 7 and Cytokeratin 20 in Primary Urinary Bladder Carcinoma and Matched Lymph Node Metastasis.
Vertebral Metastases From Pineoblastoma.
NEW TEST MAY PREDICT BREAST CANCER METASTASIS.
Prostate cancer linked to defects in a single cell.
Surgery not necessary for most metastatic colorectal cancer patients.
Should we treat lung metastases from adenoid cystic carcinoma of the head and neck in asymptomatic patients?
New computer program uses artificial intelligence to diagnose metastatic cancer.
Small cell carcinoma of the lung with mandible metastasis/ Mandibula metastazli kucuk hucreli akciger karsinomu.

Terms of use | Copyright © 2014 Farlex, Inc. | Feedback | For webmasters