Introduction
Meningiomas are the second most common primary tumors affecting the central nervous system, accounting for more than 35 % of the primary brain tumors in adults [
1]. The vast majority of meningiomas are slow-growing, benign (non-cancerous) tumors, although certain subtypes are more aggressive than others, and benign does not mean that they are without risk. Depending on its size and location, a benign meningioma can cause significant problems to the patient, become life-threatening, and be extremely difficult to treat. Originating from the dural covering of the brain, meningiomas are classified based on the site of origin, the involvement of adjacent tissues (venous sinuses, bone, brain, and nerves), and histological grade. The majority of meningiomas fall into histopathological classification WHO grade I, whereas atypical meningiomas (WHO grade II), and malignant or anaplastic meningiomas (WHO grade III) occur less frequently and constitute approximately 5–7 % of all meningiomas [
2]. Being diagnosed with an intracranial meningioma means a reduction of life-expectancy to the patient, which is largely caused by tumor recurrence or outgrowth of tumor remnants, irrespective of the WHO grade. Even tumors that are (reported to be) resected completely may recur, which makes the operative treatment of every intracranial meningioma a challenge for the surgeon.
Incompletely resected tumors and high-grade lesions are frequently additionally treated with fractionated radiotherapy or stereotactic radiosurgery, and sometimes with hormonal- and immunotherapy [
3‐
7]. Recurrences that occur after adjuvant/additional treatment often leave the surgeon with reoperation as the final and only option left for the patient. Computer-assisted operative systems (neuronavigation) have the potential to assist the surgeon in maximizing the extent of surgical resection, but there are limitations of this technique due to the phenomenon of “brain shift,” and the absence of real-time feedback [
8‐
14]. To aid the intraoperative discerning of tumor from normal brain tissue, dural vessel involvement, bone invasion, and brain invasive growth, real-time fluorescence imaging was introduced, aiming to overcome the aforementioned problems. Although the first attempt of visualizing brain tumors using a fluorescent dye (fluorescein) was performed in 1948 by Moore and Peyton [
15], it took almost 50 years until the principle was applied in a larger surgical setting [
16]. Currently, 5-aminolevulinic acid (5-ALA) is used to aid intraoperative identification of malignant gliomas [
17]. Since its introduction, 5-ALA fluorescence has been applied in a range of different types of CNS tumors, such as ependymomas, hemangioblastomas, metastatic brain tumors, and also in intracranial meningiomas. A number of authors have tried to outline the merits of 5-ALA in meningioma surgery, and have published their experience in this field [
18‐
24], advocating its use in current neurosurgical practice. Nevertheless, the reliability of 5-ALA in meningioma surgery has not been established yet. Items such as specificity and sensitivity have not been solved, and the influences of histopathological grade, previous treatment like radiotherapy, radiosurgery, and medication (steroids, hormonal treatment, and chemotherapy) on the ability of meningioma cells to produce fluorescence remain unknown. In this review, we summarize the reports and clinical studies that have been published about the application of 5-ALA in intracranial meningioma surgery, and we report two illustrative cases of our institutional experience, in order to define the status quo of this technique in current intracranial meningioma surgery.
5-Aminolevulinic acid
5-Aminolevulinic acid is an indirect fluorophore and a natural biochemical progenitor of hemoglobin. Administered orally, 5-ALA is resorbed through the upper intestine into the blood, where it passes the blood–brain barrier [
25].
It provokes the synthesis of protoporphyrin IX (PpIX) and is considered an endogenous photosensitizer. Excited with fluorescence excitation light (400–440 nm), PpIX emits red light energy of 635 nm, which is visible for the human eye. Although 5-ALA accumulates fairly selective in neoplasms, this process is not entirely tumor-specific as non-malignant tissue, i.e., brain parenchyma, the subependymal zone, and choroid plexus also show PpIX accumulation [
26‐
28]. Vice versa, several studies and our institutional experiences show that not all tumors or tumor parts become fluorescent with 5-ALA [
18,
20,
21,
23]. Moreover, the degree of heterogeneity in fluorescence for different tumor grades remains a major concern [
29].
Discussion
In this review, we summarized the application of 5-ALA-induced fluorescence as an intraoperative tool to maximize the resection of intracranial meningiomas. Furthermore, we discussed our institutional experiences with this technique presenting two illustrative cases.
On aggregate, a total number of 126 patients (not including our own two cases) with 5-ALA-assisted intracranial meningiomas resection are reported. All authors used a modified surgical microscope for fluorescence-guided visualization of the tumor and followed the protocol presented by Stummer et al. for the application of 5-ALA [
17]. This protocol was designed for malignant glioma surgery and validation for meningioma surgery has not been done, as far as we know. We question the reliability of 5-ALA in intracranial meningioma surgery. As we showed, histologically identical meningiomas did not respond equally to 5-ALA administration. Furthermore, we observed a change in fluorescence response over time and a variation of fluorescence in different areas of the same tumor, which might have been caused by the influence of additional treatment (like radiotherapy or previous medical treatment) prior to reoperation of a recurrent meningioma.
Most studies [
18‐
21,
24,
30‐
32] reported fluorescence of the main tumor mass with high sensitivity and specificity. In the majority of cases, fluorescence did lead to an extension of the resection and to the removal of additional tumor.
However, only Morofuji et al. [
19] were able to consistently correlate resected tissue with histopathological findings. In three studies [
22,
23,
33], state-of-the-art fluorescence imaging with 5-ALA was compared to quantitative probe fluorescence, resulting in a higher accuracy and sensitivity of the latter technique. However, the superiority of this approach above naked-eye evaluation of 5-ALA and its implementation in the operative routine need further investigation.
In meningioma surgery, there is no ultimate evidence regarding the appropriate resection margin to minimize recurrence [
34‐
37]. Fluorescence imaging may help to more accurately determine whether excision margins are free of tumor cells. This requires validation of the exact detection threshold of the current intraoperative techniques. The findings in this review reveal that 5-ALA-induced fluorescence seems a promising approach for the detection of (remnant) meningioma cells. However, it lacks sufficient reliability regarding specificity [
18,
21,
33] and histopathological accuracy to establish a consistent real-time assistance in meningioma surgery. In some cases, where fluorescence was reported, the difference between tumor and background fluorescence is hardly detectable with the naked eye [
21,
24].
Our two illustrative cases revealed that tumor fluorescence under 5-ALA can be inconsistent and inhomogeneous, which limits the (presumed) benefits of fluorescence-guided surgery [
18,
20,
21,
23,
30‐
33]. Pfisterer et al. [
38] suggest that the genetic regional heterogeneity, that is often found in meningiomas, can be of high prognostic and diagnostic value. This could be an explanation for the inconsistency in fluorescence. Another explanation for the unspecific fluorescence pattern was provided by Masubuchi et al. [
28], who demonstrated that PpIX is quickly excreted by meningioma cells into the extracellular space. Thus, depending on the threshold of the optical systems, non-malignant tissue might show false-negative fluorescence.
As stated previously, PpIX emits visible red light energy of 635 nm upon previous excitation. The eyes threshold for detection of this light depends on the retina, and its sensitivity/specificity for red light from, e.g., inborn retina diseases and age. These physiological limitations help to understand the different interpretation in visual fluorescence and emphasize the necessity for a more objective measurement of fluorescence (by quantitative biomarkers). However, even with the use of quantitative probe, the detection of tumor will depend on the concentration threshold for PpIX and on the experience of the surgeon.
In a recently published letter, Wilbers et al. [
39] discuss their experience with an atypical meningioma. Using 5-ALA-induced fluorescence, the authors have been able to detect residual infiltration of the tumor in the dura and adjacent brain tissue. This can be essential regarding prognosis, regardless of the meningioma type. Moreover, the authors indicate the potential use of PpIX detection in adjuvant photodynamic therapy (PDT). This is worthy of further investigation as deep penetration of visible light (400–750 nm) is impeded, mainly due to autofluorescence by surrounding tissue and absorption by hemoglobin.
It is widely known that radiation therapy (RT) of the upper body, either as a direct effect of treatment or by incidental exposure, has an etiological effect on the development of brain tumors. Moreover, it is found that radiation-induced meningiomas express more atypical features and multiplicity, compared to non-radiation-induced meningiomas [
40‐
43]. Our own experience is that recurrent meningiomas that have been treated with radiotherapy can show alteration in fluorescence. Most of the studies reviewed do not mention previous radiotherapy of their patients (see Table
1). However, such information is essential for the correct interpretation of 5-ALA-induced fluorescence. Notably, it is especially this category of patients (multiple craniotomies and radiotherapy for recurrent meningioma) that urges for the development of a tumor-specific tool to improve intraoperative tumor detection and a radical resection.
Based upon the above dilemmas, the question is how to proceed.
It is evident that there is no(t yet a) solid basis for the reliable application of 5-ALA in meningioma surgery. In our opinion, 5-ALA-assisted tumor resection should be protocolled and considered for all patients with primary and recurrent meningioma. Close and detailed follow-up is needed, also taking into account the complete patient history and previous medical treatment including radiation therapy and/or stereotactic surgery. Tumor specimen should be incubated and cultured, to allow for future tumor cell research and for the development of a tumor-specific intraoperative fluorescence techniques, as recently also has been suggested by Behbahaninia et al. [
44].
Targeting of a tumor-specific biomarker with a fluorescent probe yields a high discrimination ratio between tumor and healthy tissue and as such, this technique has the potential to accurately identify tumor deposits and may be useful in the evaluation of tumor margins. 5-ALA (400–440 nm) operates in the visible light spectrum (400–750 nm); wavelengths in which the signal is partly limited by autofluorescence of background tissue and by absorption by hemoglobin. In contrary, tumor-specific imaging using fluorescent dyes in the near-infrared (NIR) range (750–1000 nm) yield better signal-to-background ratios [
45]. Van Dam et al. [
46] have successfully reported tumor-targeted fluorescence imaging in patients with ovarian cancer to detect metastatic tumor tissue. In this concept, it is essential to determine which biomarkers are upregulated most in the tumor cells. Those markers can serve as a target for a fluorescent imaging agent. Upon excitation by a laser, the reflected signal by the fluorescent agent can be detected by a sensitive camera system. Thus, identifying meningioma-specific biomarkers may be a challenging focus for future investigations. Image-guided resection of meningiomas could benefit from this concept. Fluorescent dyes with a NIR emission wavelength are likely to be more suitable for intraoperative imaging purposes due to their tumor-specificity. Thus, a tumor-specific approach would allow a more sensitive tumor delineation and assessment of tumor margins.
Conclusion
It is challenging to draw definitive conclusions regarding the role of 5-ALA fluorescence-guided meningioma surgery from the present review. 5-ALA-assisted resection does not replace the resection of tumor under white light. Moreover, this new technique serves as additional information upon the surgeon’s experience, and his/her tactile and visual perception.
The following is a list of findings, based on our institutional experiences, and the status quo of this technique in the literature.
-
Tumor cell fluorescence can appear in benign meningiomas (WHO grade I) as well as in high-grade lesions (WHO grade II and III) meningiomas.
-
There seems to be no correlation between fluorescence intensity and proliferation rate (MIB-1 Labeling Index) or mitotic index (MI).
-
Sensitivity and specificity rates vary between studies, but are generally high.
-
Within the same tumor, fluorescence can be very different (heterogenic fluorescence), showing highly fluorescent parts, and parts that do not respond at all.
-
The fluorescence response to 5-ALA administration in a recurrent meningioma of the same histological grade can differ from the response observed during the former resection, changing from “5-ALA positive” to “5-ALA negative,” or vice versa.
This review shows that 5-ALA, as a tool to guide resection of intracranial meningiomas, is still very experimental and should only be used in protocolled prospective studies. The superiority of 5-ALA-assisted resection of intracranial meningiomas regarding progression-free survival needs to be investigated in prospective cohort studies. However, the principle of fluorescence as a real-time method to assist the surgeon in achieving complete resection (and cure!) of benign intracranial tumors is very appealing. Besides, from the application of 5-ALA, the use of other real-time modalities, especially tumor-specific intraoperative fluorophores, are very worthy to be investigated.
Kwan Park, Doo-Sik Kong, Seoul, Korea
To date, the use of 5-ALA has been limited to the intra-axial cerebral neoplasm. This is the first study of review about the usefulness of 5-ALA-induced fluorescence for the intracranial extra-axial tumors. The authors reviewed the recent literatures regarding the application and the efficacy of 5-ALA for intracranial meningiomas in detail and summarized their experience. According to the authors’ review, they found that 5-ALA can be efficaciously used under the specific conditions such as recurrent cases or bone-invading cases. However, through this review, the authors have still questions about the efficacy of 5-ALA-guided surgery for intracranial meningiomas because it has heterogenic findings over time and location. Agreeing to their conclusions, the readers should acknowledge there are many criticisms about the use of 5-ALA for intracranial meningiomas.
Hans-Jakob Steiger, Düsseldorf, Germany
On the basis of personal experience with 5-ALA fluorescence during resection of cranial meningiomas, Motekallemi and coauthors reviewed the actual literature regarding the application and usefulness of the method for meningiomas. Although a number of studies regarding the issue have been published since 2007, some questions remain. Though it appears that intensity of fluorescence correlates with WHO grade, the correlation seems to be less strong than with astrocytoma. A further open issue is whether more radical resection can be achieved when using 5-ALA fluorescence. Long-term studies are necessary to evaluate a possible impact on recurrence and overall survival.