δ-Aminolevulinic acid-induced fluorescence-guided resection of brain tumors
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.156275
Source of Support: None, Conflict of Interest: None
Maximal resection of gliomas is the current standard of care. Various technical adjuncts facilitate this. Aminolevulinic acid (ALA)-induced fluorescence guided resection (FGR) is one such strategy. We review the current literature related to ALA FGR. It is based on the selective uptake of ALA into glioma cells and its preferential conversion to protoporphyrin IX. This selectivity provides a high positive predictive value for ALA induced fluorescence. Since the introduction of this technique, clinical experience supports its efficacy in improving resections in malignant gliomas when compared to other contemporary intraoperative imaging strategies such as the magnetic resonance imaging (MRI) or the adjuncts that exhibit passive permeability like fluorescein. Future research into the understanding of the basis of ALA metabolism in glioma cells and advances in visualization technology will potentially improve the scope of application of this technique.
Keywords: Aminolevulinic acid; fluorescein; fluorescence-guided resections; gliomas; review
Malignant gliomas are the most common primary brain tumors in adults, and, unfortunately, the most challenging to treat. Multimodality therapy (surgery and adjuvant chemo-radiotherapy), though not curative, has provided a ray of hope by extending, albeit modestly, the disease-free interval and the overall survival.  Surgical resection remains the mainstay of treatment and has been shown to be related to better outcomes.  This has to be achieved without compromising neurological function. Surgical management of brain tumors, and specifically gliomas, has evolved over the last decade. Improved and more sensitive imaging techniques have revealed extensions of tumor hitherto "invisible;" and advanced surgical adjuncts allow neurosurgeons to safely resect these lesions. With better understanding of the tumor biology and growth has also come the realization that, despite the advanced intraoperative imaging tools at our disposal, true "complete" resection of malignant gliomas remains elusive. It reinforces the aphorism "The more we know, the more we know how much we do not know." The contrast-enhancing boundary is not a true reflection of the tumor extent. However, it remains the most reliable and reproducible surrogate marker for delineating the tumor extent. Mounting scientific evidence testifies that radical resections improve prognosis. Hence, the goal for the neurosurgical oncologist remains a "Safe Maximal Resection." Intraoperative resection-control adjuncts can broadly be classified into "Intraoperative imaging adjuncts" that utilize conventional imaging modalities to provide two- and three-dimensional structural imaging; and "Intraoperative visualization tools" which exploit the optical phenomena to improve the intraoperative delineation of tumor by direct inspection. The former includes an intraoperative ultrasound (IOUS), intraoperative magnetic resonance (IOMR) and less frequently, intraoperative computed tomography. Although both the US  and MR , have been shown to be useful, they have their individual limitations. These limitations include the cost of the equipment, the strain on resources (consumables and manpower) as well as the time expended in utilizing these adjuncts during surgery. The extent of the tumor delineation is also dependent on the accuracy of the imaging technique chosen. Furthermore, they are unable to provide truly real-time information and need frequent intraoperative imaging updates to ascertain the status of the residual disease.
The other set of adjuncts includes optical imaging strategies such as fluorescence imaging, Raman spectroscopy, and optical coherence tomography. The present review focuses on fluorescence-guided resections (FGRs) and specifically, delta-aminolevulinic acid (ALA)-induced FGR, which is presently the most widely used application of fluorescence imaging in tumor surgery.
Fluorescence is the ability of certain chemical substances to absorb light energy (of a specific wavelength) and reflect it almost simultaneously as a characteristic wavelength (different from that of the incident light and usually of a higher wavelength). When the emitted light is within the visible spectrum of light, the resultant fluorescence can be appreciated by visualization as a specific color. All human cells contain molecules which become fluorescent when excited by light of a suitable wavelength. This fluorescence emission, arising from endogenous fluorophores, is an intrinsic property of cells that is known as auto-fluorescence. Intracellular autofluorescence is often dominated by the reduced pyridine nucleotides [NAD (P)H] and the oxidized flavins (flavin mononucleotide, flavin adenine dinucleotide), both of which are potentially useful as cellular metabolic indicators. In tissues, the extracellular matrix usually contributes to tissue autofluorescence more than the cellular components owing to the relatively higher quantum yield of collagen and elastin among the endogenous fluorophores. Autofluorescence is often weak and nonspecific and not clinically useful at least in brain tumors. Frequently, the molecules of interest are nonfluorescent or the fluorescence obtained from them is inadequate. The targets, therefore, need to be labeled using extrinsic probes or fluorophores. The exogenous fluorophores can be broadly classified into two main groups: Fluorophores synthesized in tissues after administration of a precursor molecule, like protoporphyrin IX (PPIX) fluorescence obtained on exogenous administration of 5-ALA; and fluorophores administered as exogenous drugs, including fluorescein, indocyanine green, and photosensitizers used for photodynamic therapy (PDT) such as hematoporphyrin derivative and tetra (m-hydroxyphenyl) chlorin.
An "ideal" fluorescent marker for in vivo diagnostics should have a tumor selective uptake with a high tumor-to-normal ratio. It should be retained in tumor tissues long enough so that intraoperative imaging is possible over a substantial period of time. The marker should be biocompatible and nontoxic so that it should be administrable systemically without causing any adverse effects. A fluorescent marker that has several, but not all, of the above properties is 5-ALA induced PPIX. For this reason, it has been investigated extensively in the literature.
Uptake of aminolevulinic acid by brain and glioma tumor cells
Aminolevulinic acid does not normally cross the intact blood-brain barrier (BBB). ,, In malignant gliomas, due to the breakdown of this BBB, ALA is able to cross it. Once ALA enters the extracellular compartment, it needs to cross the cell membrane. This uptake may be passive at low concentrations but uses an active transporter at higher concentrations of ALA. , The exact mechanism by which this happens is unclear. Peptide transporter 2 may be involved. ,, Upregulation of this transporter mechanism may contribute to the relative increase in ALA uptake by tumor cells.  However, there is indirect evidence to suggest that there may be other mechanisms for the selective uptake of ALA that are as yet not elucidated. ,
Metabolism of aminolevulinic acid
Aminolevulinic acid is a normal precursor of heme utilized by virtually every cell of the body. In most normal cells, synthesis of ALA (by ALA synthase) is the rate-limiting step. Depending on the requirements of the tissue, the end-products (protoheme and heme) exert a negative feedback on further conversion of ALA, thus keeping the cycle in check. Even exogenously administered ALA is unable to overwhelm this checkpoint. In glioma cells, however, there seems to be a loss of this regulatory mechanism, thereby allowing more ALA to enter the cycle. Furthermore, the conversion is arrested at the level of PPIX. One of the reasons for this selectivity of glioma cells for ALA is the relative deficiency of ferrochelatase enzyme in them. Indeed, silencing of this enzyme has been shown to increase the accumulation of PPIX in tissues. , Ohgari et al. have shown that a balance between production of heme cycle intermediates and catabolism determines the levels of PPIX.  Expression of frataxin which is dependent on p53 may be a major determinant in this pathway.  However, there could be multiple inter-related mechanisms at work. The resultant accumulation of PPIX can then be exploited for visualization using suitable microscope filters.
Physics of aminolevulinic acid-induced fluorescence
The net result of administration of ALA in malignant gliomas is the accumulation of PPIX.  PPIX is an endogenous fluorophore. When this is excited by blue light (370-440 nm), it emits a red fluorescence (635 and 704 nm). A suitably modified operating microscope can facilitate visualization of this fluorescence and thereby guide the surgical resection. 
The basis of ALA-guided FGR is the specific intracellular accumulation of fluorescing PPIX. This has been demonstrated in preclinical in vitro and in vivo studies.  Fluorescence has been shown to correlate with the grade of the tumor.  The degree of fluorescence corresponds to the concentration of PPIX in the tissues.  This has also been shown to be related to the concentration of mitochondria within the cells. 
In clinical studies too, strong fluorescence has been shown to have a very high positive predictive value (PPV) for solid, cellular and mitotically active tumor; the value being in the range of 95-100%. ,,,, In a recent meta-analysis, Zhao et al confirmed consistently high positive likelihood ratios.  Our experience was similar (unpublished data) with strong fluorescence having a very high PPV (94% for detecting coalescent solid tumor and 100% for detecting all tumor tissue). Occasionally, falsely fluorescing biopsies have been reported. One of the reasons for the false positive result could be the reactive changes in the brain or invasion of the brain with inflammatory cells, especially in the microscopic vicinity of tumors. , Therefore, it must be remembered and reinforced here that though the fluorescence will most of the time accurately reveal the tumor extent, the decision to resect it has to be based on sound neurosurgical acumen, especially in eloquent areas (in which infiltrating tumor cells will possibly be found in the functionally relevant brain). Furthermore, though the PPV has been consistently found to be high for tissues emitting a strong fluorescence, it is not so for "vague" or "weak" fluorescence. This means that areas especially at the edges (where the fluorescence pattern is not as strong), may not always represent solid tumor. More often, it is the infiltrating, less cellularly dense, tumor edge. Visualizing fluorescence using a surgical microscope has limitations in terms of the sensitivity of the fluorescence that it can detect. It has been shown, using far more sensitive techniques such as spectroscopy and confocal microscopy, that even diffuse, low-grade gliomas accumulate PPIX and fluoresce. , It is just that using the microscopic filters presently available for routine clinical use, this fluorescence is not detectable. This is because the optically perceived fluorescence is actually particularly strong in tissues. Hopefully, future advances will allow application of this technique in low-grade, diffuse gliomas.
Though most studies have shown consistency in the reporting of PPV, the results for negative predictive value (NPV) and sensitivity (Sn) and specificity (Sp) have been variable. As described recently, one of the reasons for this is the methodology by which such studies may have been conducted.  For NPV, Sn and Sp, it is essential to have adequate sampling from nonfluorescing and normal areas to ensure adequate true and false negatives. This may not have been routinely done (or it may not even have been possible) in all these studies. However, as a diagnostic tool, a high PPV is reassuring to the surgeon, as it means that whenever strong fluorescence is encountered, the likelihood of it representing the tumor tissue is high.
Typically, a contrast enhancing high-grade glioma demonstrates a central nonfluorescing necrotic core, surrounded by a rim of solid, strongly fluorescing viable tumor of variable thickness [Figure 1]. Toward the edge, the fluorescence fades away in the infiltrating zone. Occasionally, nonspecific ependymal fluorescence (of uncertain significance) may be detected. 
Aminolevulinic acid FGR is most suited for contrast enhancing tumors. Stummer et al.  initially showed that the intraoperative residual fluorescence correlated with postoperative contrast enhancing residual disease. This association was better when the quality of residual fluorescence was reported as strong. Roberts et al. have meticulously shown that a strong fluorescence correlates well with the contrast enhancing tumor component.  They used coregistration of the MR with the intraoperative fluorescence and correlated the biopsies with histopathological examination. Though contrast-enhancing tumors are more likely to fluoresce, the fluorescence extends beyond the contrast enhancing borders. Schucht et al. used an indirect method of calculating tumor volume, and demonstrated that using ALA, on an average, they were able to resect 45cc of additional ALA positive, but noncontrast enhancing tumor.  The significance of resecting this "extra" tumor volume has been shown in another study by Aldave et al.  In this study, the authors analyzed a group of patients who had complete resection of the enhancing tumor (CRET) using ALA. In some of them, however, there was intraoperative residual fluorescence (though postoperative MR did not reveal any enhancing residual tumor). These patients had poorer median overall survivals (17.5 months vs. 27 months) than the patients where all fluorescing areas were resected, a difference that persisted even on multivariate analysis.
In this context, it must be said that positron emission tomography (PET), using labeled amino acid tracers, may correlate better with the true extent of the tumor. A recent systematic review concluded that fluoro-ethyl-tyrosine PET (FET PET) can reliably diagnose brain tumors.  ALA-induced fluorescence has been shown to correlate well with the extent of the tumor delineated by FET PET.  In glioblastomas, ALA fluorescence has been shown to delineate the tumor boundary to the maximum extent (even beyond the FET PET and certainly beyond the contrast enhanced MRI).  Even in diffuse gliomas (that can be histologically heterogeneous), it has been shown that FET PET is very sensitive in picking up "hot spots" which represent "higher grade" areas. These FET PET hot spots show a strong correlation with focal areas of ALA fluorescence whereas contrast-enhanced MRI has a poor correlation.  A similar experience has been reported for ALA in correlation with 11 C methionine PET. 
Stummer et al. described one of the first large clinical studies of ALA in malignant gliomas.  In this study, MR-documented complete resections were achieved in 63% of 52 glioblastoma multiformes (GBMs). This study demonstrated that ALA FGR was useful in resection of malignant gliomas. It also showed that residual fluorescence correlated with adverse survival. This study was followed by the seminal randomized trial published in 2006.  This study compared patients operated using ALA-FGR versus routine microscopic white light-guided surgery. 270 patients (139 in the ALA arm and 131 in the white light arm) were analyzed. The extent of resection was volumetrically measured in all cases. The study demonstrated that using ALA-FGR, higher rates of gross total resection (65% vs. 36%) were achieved. Although the study was not powered for survival, it did demonstrate doubling of progression-free survival (41% vs. 21%) in the ALA FGR group. Moreover, this result was achieved without any significant increase in functional deficits and was associated with negligible toxicity related to the use of ALA. In a subsequent post-hoc analysis, they also found that across the two groups, radical resection rates were associated with better survival.  This underlines the importance of gross total resection (irrespective of the adjunct used); but also proved that this increased gross total resection was more often possible using ALA. One must, however, remember that the patient sample for this study consisted of subjects having predominantly enhancing, potentially resectable malignant gliomas, and as such, the results cannot be extrapolated to all malignant gliomas. Having said that, even in these so-called potentially resectable gliomas, conventional strategies have yielded suboptimal resection rates (as is evident in the control arm of the study itself), and the use of ALA FGR can certainly benefit these patients. It is worthwhile to remember that the original study by Stummer mentioned above was published almost 10 years ago. For most participating surgeons, these were the initial few cases they would have operated using ALA. More contemporary series have shown that with experience, much higher gross total resection rates can be achieved. Diez Valle reported CRET rates of 83% in a cohort of 36 patients, with a mortality of 0% and a morbidity of 8.2%.  In another recent series, Schucht et al. reported CRET in 89% cases of GBMs.  Using the volumetric MR criteria (as was done in the original ALA study), their gross total resection rates were actually higher at 96% in the "potentially resectable" group. In our experience too (unpublished data), we could achieve 75% complete resection rates. Thus, as is true of any technology, there is always a learning curve and after gaining sufficient clinical experience, optimal results can be achieved.
Since 2005, the Stupp regimen for adjuvant therapy in glioblastomas has become universally accepted (which was not a part of the routine adjuvant treatment during the period when the ALA study was conducted), contributing to improved overall survival in glioblastomas.  In the Stupp study (when it was further stratified for the extent of resection), patients who had undergone gross total resection, did better. Thus, the use of ALA to improve gross total rection rates could potentially contribute positively to the overall survival in the currently practiced clinical situation. However, direct extrapolation of results of clinical trials (which have a very selected sample and are performed in very controlled conditions) to the entire patient population does not usually yield similar results. Therefore, continued reporting of data from routine practice settings is essential to gauge the potential clinical impact of an intervention. In a retrospective observational study across Spain (the VISIONA study), the use of ALA routinely during malignant glioma surgery was found to yield better resection rates (67% vs. 45%) and an improved 6-month progression-free survival (69% vs. 48%).  This benefit persisted even after controlling for other known predictors of outcome.
One must remember that while using ALA, attention to basic surgical techniques (see Practical Tips) is imperative to ensure minimization of errors and maximization of the benefits of the technique. Schucht et al. reported that in around 6% of cases, while using ALA, there were unanticipated residues.  They reoperated these cases and in more than half, found residual fluorescence hidden behind layers of nonfluorescent tissue/debris. Thus, it must be remembered that ALA is a tool to be used judiciously and in concert with all other available adjuncts to ensure the best outcome for the patient.
One of the concerns with the use of ALA FGR (and for that matter with any other adjunct that potentially improves the radicality of resections) is the risk of added neurological morbidity. However, the extended resection with ALA guidance does not come at the cost of increased morbidity.  Optimizing intraoperative resection is essential to avoid complications. In fact, patients with safely performed extended resections are more likely to remain neurologically stable for longer periods of time. In the original ALA study, patients in the ALA arm were less likely to require resurgeries for recurrences/tumor progression.
Even in tumors close to eloquent areas, it is possible to obtain radical resections using ALA combined with functional monitoring techniques. Della Puppa et al reported CRET rates of 74% using ALA in eloquent region gliomas.  In all the remaining 26%, where the intraoperative neurophysiological monitoring necessitated premature termination of resection, they could still resect at least 90% of the tumor. Similar results have been described in other studies also. , In our experience (unpublished data), in eloquent region tumors, the complete resection rates were 66%.
Safety profile of aminolevulinic acid fluorescence-guided resection
In the randomized study by Stummer et al., there was no difference in the toxicity profiles of the two groups. In most of the subsequently reported studies, no significant toxicities have been reported. , Even in a recent large review of 207 cases, the safety profile was very favorable with no reported morbidity attributable to the use of ALA.  However, one must be cognizant of the potential complications and implement appropriate protocols to ensure minimization of complications (See Practical Tips section). Known cases of porphyrias are an evident contraindication for the use of ALA, although no cases have been reported that have been treated with ALA. Direct exposure to sunlight (especially to light in the red wavelength) has to be avoided for 24-48 h after ALA exposure in order to avoid photoxicity, which though not alarming, has been occasionally reported.
It is very important to understand that ALA FGR is one important tool at the neurosurgeon's disposal during tumor resection. It needs to be used judiciously and often in conjunction with other adjuncts. Each tool has its own advantages and disadvantages. The intelligent neurosurgeon is able to combine them so as to obtain the maximum combined benefit for the patient. ALA FGR is essentially a surface visualization technique providing real-time information. On the other hand, intraoperative MR or US provide cross-sectional imaging. In practice, combining both techniques should provide synergistic benefits.  In certain situations, one technique may be better than the other. When compared to IOMR, it was found that for contrast-enhancing tumors, ALA FGR was better than IOMR; whereas for nonenhancing tumors, IOMR was more beneficial.  Similar experiences have been reported for ALA and IOUS too.  However, direct comparison between ALA and IOMR or IOUS is fraught with technical challenges and difficulties. In a meticulously conducted study, Coburger et al compared the diagnostic accuracy of ALA and IOMR (postcontrast) in detecting tumor tissue. They found that IOMR was as good as ALA for the solid tumor component; however, the infiltrating edge was better detected using ALA.  This provides factual proof and corroborates similar clinical experiences reported earlier where ALA was found to detect and permit resection of the tumor well beyond its extent detectable by contrast enhancement. ,
The biological basis of ALA induced fluorescence and its relative specificity for viable tumor cells opens exciting avenues for translational research in neuro-oncology. Piccirillo et al have shown that ALA induced strong fluorescence can be used as a reliable surrogate marker for a proliferative tumor, permitting appropriate sampling of geographically divergent areas within the tumor.  In another similar correlative study, it has been demonstrated that the variably fluorescing zones within glioblastomas are made up of cells of different biological characteristics.  The intratumoral heterogeneity may also be revealed by the fluorescence.  It is well-known that glioblastomas are extremely heterogeneous phenotypically as well as genotypically. The fluorescence patterns may reflect intratumoral heterogeneity which may be further investigated to reveal basic fundamental processes in tumor pathogenesis.
Though the main approved indication for use of ALA FGR is in malignant gliomas, it has been tried in a range of other tumors.
The effects of prior treatment as well as surgical scarring may influence the utility of intraoperative fluorescence in the case of recurrent tumors. Prior surgery alone may not be of much concern as was reported in a recent study.  However, effects of prior radio-chemotherapy may adversely affect the accuracy of ALA induced fluorescence. That notwithstanding, Nabavi et al. reported a high PPV (93-97%) of ALA FGR in recurrent gliomas.  However, false positive results due to the presence of reactive changes are a potential worry, and this must be borne in mind while employing ALA FGR during surgery for recurrent gliomas.  As was the case during the first surgery, the limits of resection imposed by the eloquence of the region in the vicinity, need to be respected.
As mentioned earlier, most low-grade diffuse astrocytomas do not show visual fluorescence. However, using more sensitive techniques such as spectrometry  and confocal microscopy,  it has been shown that even these low-grade gliomas accumulate protoporphyrin. Clinical detection with currently available tools is not reliable, and hopefully in the future, better tools will be available. Even so, ALA may still be useful to guide sampling within these so-called diffuse low-grade tumors so as to ensure correct targeting of anaplastic foci and avoid under-grading of the tumors. 
Besides gliomas, the use of ALA has been sporadically reported in metastases, meningiomas, medulloblastomas, pituitary adenomas, hemangioblastomas, various spinal tumors, and a range of pediatric brain tumors. ,,,,,,, Endoscopic visualization of ALA-induced fluorescence has also been described and could be a useful addition to microscopic visualization, especially in difficult-to-reach areas. 
Besides ALA FGR, another technique of FGR currently being evaluated for brain tumors uses fluorescein. This technique, however, is not entirely new. The first use of fluorescein in neurosurgery was actually in 1948.  Fluorescein sodium is a strong water-soluble fluorochrome of low molecular weight that fluoresces strongly in the green and yellow range. The compound appears safe for intravenous application; however, anaphylactic reactions have been reported.  After injection, fluorescein has a plasma half-life of 23.5 min. 80% is converted to a monoglucuronide within 1 h which is similarly fluorescent. The half-life of the monoglucuronide is 264 min. Elimination of fluorescein is prolonged with a duration of 48-72 h after administration of 500 mg. , Thus, a prolonged plasma fluorescence can be observed after administration, and tissues will fluoresce under appropriate excitation wherever there is intact perfusion. In malignant brain tumors with their inherent blood-barrier breakdown, fluorescein is extravasated into the tumor tissue, thus appearing to provide a marker for tumor tissue. Fluorescence intensity and dynamics in this situation depend directly on blood volume and perfusion in these tissues, as well as on extravasation. If the blood-brain barrier is disrupted, fluorescein will readily diffuse into the brain tissue. Thus, in malignant brain tumors, there will be sequestration of fluorescein into the edematous extracellular tumor space and some association with tumor tissue may be noted.
After this initial description, there is very scanty information on the use of fluorescein until the late 1990s when the works by Kuroiwa et al was published, , with the last publication appearing in 1999. Today, more than a decade later, fluorescein sodium is again under scrutiny for FGR of malignant gliomas, resulting in a number of papers describing its use in malignant glioma surgery , and especially after the introduction of a new filter system for the last generation of Zeiss surgical microscopes, the Yellow 560 filter (Zeiss, Oberkochen, Germany). There are a number of theoretical advantages of fluorescein sodium over 5-ALA when assessed using this filter. Fluorescein can be injected at any time, whereas 5-ALA has to be given orally several hours before surgery. Fluorescein, as used in ophthalmology, has a lower cost compared to 5-ALA. The Zeiss Yellow 560 long pass filter is conceived to permit passage of more visible light than the Zeiss Blue 400 long pass filter for visualizing the red PPIX fluorescence. Thus, the user's impression is yellow-green fluorescence on an almost normal background, whereas with the Blue 400 filter the surgeon sees red fluorescence on a blue background. On the other hand, while 5-ALA is associated with glioma cell metabolism and has a proven high positive predicting value (as described earlier), fluorescein is a passive marker of BBB disruption and is found in all tissues which are perfused, as well as in cerebrospinal fluid and blood. If blood contaminates the surgical cavity, the exposed tissues will be marked as well (Schwake et al 2015 in press). This raises many concerns regarding its selection for clinical use and cautions that at present, fluorescein should not be used outside controlled clinical studies and only when appropriate degree of prudence is exercised in interpreting the fluorescence signal. At present, fluorescein is not approved for brain tumor surgery and thus its usage is off-label [Table 1].
Little is known about the best timing of tumor visualization after the intravenous fluorescein application. Injecting this fluorochrome too early before the intra-operative visualization may result in its significant unspecified propagation in regions with edema leading to a loss of accuracy in delineating tumor margins. Injections administered immediately before or during visualization might theoretically be useful for detecting perfusion abnormalities in tumor tissue and thus in the identification of residual tumor. However, this has often not been observed. In fact, fluorescein may actually miss the tumor highlighted using 5-ALA (Schwake et al 2015, in press). In addition, blood levels in the tissue are high, leading to a strong, nonspecific fluorescence of all perfused brain tissue. Overall, the way fluorescein is being used at the present moment raises concerns regarding whether it may be used safely with the purpose of selective identification of tumor tissue. Further robust clinical evidence is awaited.
Despite almost a decade since its initial description, ALA FGR has not gained widespread application in major neurosurgical centers in India. Only a handful of centers provide this facility (personal communication). Upgradation of the operative microscopes, although expensive, especially in a resource-constrained setting, is certainly possible in most major academic centers. The major hurdle, however, remains the availability and the cost of ALA. As the drug is not approved for marketing in India, its use on compassionate grounds requires licensing permission from the regulatory body (Drug Controller General of India) on an individual basis. Academic centers can apply for the institutional license for use of the drug (as has been done at our center). The prohibitive cost of the drug (approximately 120,000 INR) makes it relatively inaccessible for most patients. In the absence of widespread insurance coverage and reimbursement in healthcare, a very small percentage of eligible patients can actually afford it. Although its cost-effectiveness in the overall management of malignant gliomas has been demonstrated in Europe, , its routine use may not be applicable in the Indian setting where disparity in health care costs is wide. At some centers, the cost of ALA alone may be higher than the cost of the entire surgery (and often also the adjuvant therapy using generic drugs).
At the Tata Memorial Centre, we have been using ALA FGR since 2012 (unpublished data). We have so far used it in 28 cases achieving 75% complete resection in potentially resectable malignant gliomas and 66% complete resection in those situated in eloquent areas. In our experience, the use of ALA has not caused any toxicity.
The utility of ALA cannot be discounted; efforts to understand the cost-benefit ratio and to adopt further measures to improve the cost-effectiveness are essential. Concerted efforts in collaborative groups in a multicentric setting can provide solutions. Having a few large designated referral centers equipped with this technology may reduce infrastructural costs. The cost of the drug itself still remains an issue.
As the experience with ALA FGR increases, it is imperative to have prospective data documenting the diagnostic accuracy and predictive value of the technique across centers. This will help to improve the way ALA FGR is currently used and may possibly pave the way for expanding its indications to other tumors where its usage at present is infrequent. Further research addressing the biological basis and efficacy of the technique in a range of brain tumors is warranted. The efficacy of the technique is also currently limited by the sensitivity of the currently available microscopes and the integration of its filters. Improvements in operative visualization technology are likely to improve the sensitivity of ALA FGR. Biological manipulation of the heme pathway is a potential research avenue whereby the concentrations of PPIX can be manipulated to improve the fluorescence. ,, Finally, ALA can also be used for PDT. This involves exposure of the tissues to a slightly different wavelength of light for a longer time to obtain cytotoxic effects by virtue of production of a singlet oxygen free radical.  Currently, the technique is under experimental evaluation.
Aminolevulinic acid FGR is a revolutionary technique that permits a reliable intraoperative visualization of malignant gliomas. Its use facilitates a more radical excision and thereby improves survival. The active uptake of ALA and selective accumulation of PPIX within glioma cells makes this a very attractive tool to accurately delineate tumor cells. This has been validated in large studies. Other techniques of FGR are presently under evaluation and need validation in larger clinical trials. Further refinements in instrumentation are likely to improve the accuracy of ALA FGR.
We would like to thank Dr. Epari Sridhar from the department of Surgical Pathology for providing the photomicrographs. We also appreciate inputs from Dr. Shabbir Hussein and his help in collecting information for the review.