Editorial; Journal of Neurosurgery - = 113(2):Pages 199-201

Home =20

Activate=20 Account

Quick=20 Links=20

=E2=80=A2	Email=20 this article
=E2=80=A2	Add=20 to Favorites
=E2=80=A2	View=20 related articles
=E2=80=A2	Download=20 to Citation Manager
=E2=80=A2	= Most-downloaded=20 articles

August 2010 Volume 113, Number = 2

PDF
TOC=20 Alert
RSS=20
Table of=20 Contents=20
Past=20 Issues
JNSPGOnline=20 =

Editorial

Malignant = meningiomas

Roberto C.=20 Heros, M.D.

University of=20 Miami School of Medicine, Miami, Florida

Please include this = information when=20 citing this paper: published online March 12, 2010; = DOI:=20 10.3171/2009.12.JNS091936.

DOI: = 10.3171/2009.12.JNS091936

Introduction	ReferenceResponse

The article by Sughrue and colleagues¹ in this=20 issue of the Journal of Neurosurgery is = important for=20 several reasons: it is the largest published series of = patients with malignant meningiomas treated at a = single=20 institution (63 patients); the follow-up is relatively = long=20 (median followup time 5 years; range 1=E2=80=9322 = years); and the=20 survival results are well described using Kaplan-Meier = analysis. The most interesting result is that patients = with=20 subtotal resection (STR) followed by fractionated = radiotherapy=20 actually did better than patients in whom the = operation=20 consisted of gross-total resection (GTR) followed by=20 radiotherapy. Although this unexpected result deserves = some=20 analysis and discussion, it should not obscure the = fact that=20 there are other important findings in this paper. I = will=20 proceed to discuss the latter ones before commenting = on the=20 issue of STR versus GTR.

In my opinion, the most important finding in this = study is=20 that radical surgery (gross-total or subtotal = excision) is an=20 effective treatment for recurring malignant = meningiomas that=20 were initially treated with surgery and postoperative=20 fractionated radiotherapy. Patients who underwent a = second=20 operation on recurrence (37 patients) had a median = survival of=20 53 months after the operation, whereas those who were = not=20 treated with surgery at the time of first recurrence = (8=20 patients) had a median survival of only 25 months. = Clearly,=20 there was a significant selection bias and, as the = authors=20 admit, older patients and those with significant = neurological=20 impairment were less likely to undergo a repeat = operation.=20 However, when the authors corrected for age and = Karnofsky=20 Performance Scale (KPS) score on Cox regression = analysis,=20 there was still a major survival advantage with repeat = surgery. Importantly, such repeat operations did not = appear to=20 result in significantly increased morbidity as = compared to=20 primary operations. The fact that a second operation = is of=20 significant value for recurrent malignant meningiomas = is=20 particularly important because it is well known that = other=20 forms of salvage therapy, particularly chemotherapy, = are of=20 little value under these circumstances.

This study confirms that patients with malignant=20 meningiomas do much better after radical surgery and=20 postoperative fractionated radiotherapy than do = patients with=20 more common intracranial malignancies such as primary = brain=20 tumors and metastasis. In this particular series, the = 2-, 5-,=20 and 10-year survival rates were 82, 61, and 40%, = respectively.=20 Although the fact that malignant meningiomas have a = much=20 better prognosis than other intracranial malignancies = is well=20 known, it is nice to have that confirmed in the = largest series=20 of such tumors reported.

In this particular series, the addition of focal=20 radiotherapy, either with stereotactic radiosurgery = (SRS; 13=20 patients) or with ¹²⁵I interstitial = brachytherapy=20 (13 patients) after repeat operation on recurrence, = did not=20 seem to add a significant survival advantage. It = appears to=20 me, however, that with these relatively small numbers = and with=20 the retrospective nature of the study, this question = remains=20 unanswered. Incidentally, another important question = that=20 remains unanswered is whether focal radiosurgery at=20 recurrence, in those tumors that are small enough to = be=20 treated with this modality, could be as effective as=20 resection. Likewise, because all patients were treated = with=20 fractionated radiotherapy after the initial operation, = we do=20 not know if patients in whom gross-total excision was = achieved=20 would have done as well without radiotherapy, or if = SRS for=20 the small residual tumor in patients with subtotal = excision=20 would have been as satisfactory.

In this series, the location of the tumor appears = to have=20 made no difference in terms of postoperative morbidity = or=20 survival. This is a bit surprising, because one would=20 naturally expect patients with tumors in more critical = and=20 surgically difficult locations such as the skull base = to do=20 worse. The authors address this issue well in their=20 discussion, and their explanation is that the = operations for=20 tumors of the skull base were likely to have been less = aggressive. Why the individuals in this subgroup lived = as long=20 as patients with meningiomas in more accessible = locations,=20 whose tumors were probably removed more radically, = relates to=20 what I consider the most interesting finding of this = study,=20 and that is that patients with STR did as well as = (actually=20 better than) those with GTR in terms of survival.

There were 34 patients who underwent primary = resection at=20 the authors' institution (University of California at = San=20 Francisco [UCSF]). Of these, 23 patients had what the = authors=20 estimated at surgery and on postoperative studies to = have been=20 a GTR, and 11 patients underwent STR (which the = authors prefer=20 to call =E2=80=9Cnear-total=E2=80=9D resection, = because in their estimate,=20 these patients had at least 90% of their tumor = removed). It is=20 important to emphasize that all of these patients = received=20 postoperative fractionated radiotherapy. Clearly, the = finding=20 that the patients who had STR had a longer median = survival=20 than those who underwent GTR (107 vs 50 months) is=20 counterintuitive. Similar but less robust findings = applied to=20 the group of patients who underwent a second operation = at=20 UCSF. The authors discuss the plausible explanations = for these=20 findings well, and admit that they could not come up = with a=20 good explanation. Both patient groups had similar KPS = scores=20 before surgery. There was no difference between the = times of=20 initiation of radiotherapy in the 2 groups. It is not = clear=20 from my reading of the paper whether the authors = looked for=20 other variables such as age, preoperative = comorbidities, size=20 and/or location of the tumor, and so on, which may = have played=20 a role in the difference in survival in the 2 groups; = however,=20 I suspect that they would have commented on this if = they had=20 seen any gross differences in these factors in the 2 = groups. I=20 admit that after carefully analyzing their data and = looking=20 for a reasonable explanation for this unexpected = finding, I=20 can do no better than the authors.

The rate of serious neurological complications was = higher=20 in patients who underwent GTR (25 vs 11%), and it is = possible=20 that this difference, although not statistically = significant,=20 influenced the results. That is, patients with serious = neurological morbidities may have been more prone to = the=20 development of life-ending medical complications, = and/or were=20 treated less aggressively for their complications than = the=20 patients who were doing well neurologically after = surgery.=20 This explanation is reinforced by a quick glance at = the=20 Kaplan-Meier graph, which indicates that the deaths in = the=20 group with GTR occurred gradually over the first few = years,=20 whereas most of the deaths in the STR group occurred = later on,=20 at the time of recurrence. Still, even though this = factor may=20 at least partially account for the better survival of = patients=20 with STR, it certainly does not explain why GTR did = not result=20 in a significantly delayed time of recurrence, and = therefore=20 longer survival.

From the practical point of view, the best lesson = to take=20 from these results is, to quote the authors, that = =E2=80=9Caggressive=20 attempts at GTR in cases without a clear plane of = resection do=20 not seem to be significantly beneficial in these = patients, and=20 may be harmful in some cases.=E2=80=9D Additionally, I = think the=20 authors' data allow us to conclude that with these = malignant=20 meningiomas, to leave a small amount of residual tumor = does=20 not seem to worsen survival, provided that the patient = is=20 treated with postoperative radiotherapy. It would not = be=20 prudent and it may be harmful for neurosurgeons to = conclude=20 from these data that STR is preferable, even in cases = in which=20 GTR could be achieved with little likelihood of = additional=20 morbidity. Although the statistical analysis of these = data=20 indicates that STR is better than GTR, I submit (at = the risk=20 of sounding heretical to some of my colleagues) that = this is a=20 case where common sense should trump statistics.

In summary, this article provides substantial = support for=20 aggressive resection, both as primary therapy and at = the time=20 of recurrence, for malignant meningiomas. The data = suggest=20 that attempts to remove the last bit of tumor stuck to = eloquent brain or to cranial nerves or major vascular=20 structures result in higher morbidity, and are not = necessary,=20 because survival seems to be at least no worse in = patients=20 with subtotal excision when surgery is followed by=20 fractionated radiotherapy. This study also confirms, = in a=20 large series of patients, that with aggressive surgery = followed by fractionated radiotherapy, these tumors = are=20 associated with much longer progression- and = tumor-free=20 survival times than other types of intracranial = malignancies.=20 This study also clearly shows that an aggressive = repeat=20 operation is effective in treating recurrences of = these=20 tumors. However, the study does not address the = important=20 question of whether such recurrences, when small = enough, can=20 be treated as effectively with radiosurgery. It also = does not=20 address the possible role of radiosurgery after = subtotal=20 excision, or simple observation without radiotherapy = after=20 gross-total excision. It is unlikely that these = questions will=20 be answered with randomized studies, because I suspect = that=20 neurosurgeons, including myself, will prefer to resect = recurrences in accessible areas and use radiosurgery = only for=20 inaccessible tumor recurrences, and will continue to = use=20 fractionated radiotherapy preferentially after primary = total=20 or incomplete excision of these malignant tumors. This = policy=20 is supported by the excellent results obtained by the = authors=20 adhering to these principles.

The authors of this study ought to be=20 congratulated on an excellent analysis of this large = series of=20 patients with malignant meningiomas.

Reference	IntroductionReference <<Response

Sughrue ME, Sanai N, Shangari G, = Parsa=20 AT, Berger MS, McDermott MW: Outcome and survival = following=20 primary and repeat surgery for World Health = Organization=20 Grade III meningiomas. Clinical article. = J = Neurosurg [epub=20 ahead of print March 12, 2010. DOI:=20 10.3171/2010.1.JNS091114]

Response	IntroductionReference

We thank Dr. Heros for his detailed and attentive = analysis=20 of our work. Our interpretations of these results are = similar=20 to the views he puts forward in his editorial. = Importantly, we=20 agree with his assessment that there remain a number = of=20 unanswered questions regarding the management of the = disease=20 in patients with malignant meningiomas.

Although our experience supports the treatment = paradigm of=20 aggressive resection and external-beam radiotherapy as = initial=20 treatment, followed by repeat resection at the time of = first=20 recurrence, the question of what to do after this = remains=20 unresolved. Are additional adjuvant treatments, such = as SRS or=20 3D conformal radiotherapy helpful? This is an = especially=20 important question given the frequent need to leave = behind=20 small portions of tumor to prevent serious = neurological=20 injury. Although we did not note a survival effect for = the=20 small number of patients receiving radiosurgery or=20 brachytherapy, as Dr. Heros notes, these data are = hardly=20 definitive, and more rigorous studies are = warranted.

Another important remaining question is this: = =E2=80=9CWhat is the=20 appropriate course of action to take upon second = recurrence?=E2=80=9D=20 Given the apparent benefit of repeat resection, one = approach=20 would be to perform a third operation, if possible, = and in the=20 absence of more clearly defined, better options, this = might be=20 the best approach. It should be noted, however, that=20 experience with 3 operations is limited, and the = analysis of=20 the effect of this approach is complicated by small = numbers=20 and heterogeneous patient populations. There remains a = realistic possibility that a doubly recurrent tumor is = biologically distinct from the original lesion, and = might be=20 more aggressive, especially after having been = subjected to the=20 selective pressures of 1 or even 2 rounds of radiation = treatment. Thus, although it seems reasonable to = perform=20 resection of a second recurrence in patients with good = performance status, further data to support or refute = this=20 idea are needed.

Again, we appreciate Dr. Heros' = interest in our=20 work.

Michael E.=20 Sughrue, M.D., Nader = Sanai,=20 M.D., Gopal Shangari,=20 A.B., Andrew T. Parsa, = M.D., Ph.D., Mitchel = S. Berger,=20 M.D., and Michael W. = McDermott,=20 M.D.

University of California at San = Francisco,=20 California