Application of a Novel Miniaturized Histopathologic Microscope for Identifying Cerebral Glioma Margins Rapidly During Surgery

doi:10.21203/rs.3.rs-2448561/v1

Download PDF

Research Article

Application of a Novel Miniaturized Histopathologic Microscope for Identifying Cerebral Glioma Margins Rapidly During Surgery

https://doi.org/10.21203/rs.3.rs-2448561/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Rapidly and accurately identifying glioma margins during neurosurgery has significance in reducing the recurrence rate and improving perioperative safety. The purpose of our study is to assess the clinical performance of the DiveScope, a novel handheld histopathologic microscope in rapidly differentiating glioma from normal brain tissue during neurosurgery. Prior to the application of the DiveScope to neurosurgery, we first dissected 3 experimental pigs' brains and observed them under the DiveScope in vivo. Subsequently, 32 ex-vivo specimens from 18 patients were included in the present study. The excised suspicious tissue was sequentially stained with sodium fluorescein and methylene blue and scanned with DiveScope during surgery. The adjacent tissue was sent to the department of pathology for frozen section examination. They would eventually be sent to the pathology department later for hematoxylin and eosin (HE) staining for final confirmation. The positive likelihood ratio (PLR), negative likelihood ratio (NLR), sensitivity, specificity, and area under the curve (AUC) of the device were calculated. In addition, we compared the difference in time consumption between DiveScope and frozen sections for the initial judgment. We also provided a typical case to elucidate the clinical application of the DiveScope in neurosurgery. When HE-stained sections were used as the gold standard, the sensitivity and the specificity of the DiveScope was 88.29% (95% CI: 70.8% − 97.6%) and 100% (95% CI: 39.8% − 100.0%), respectively. In contrast, the sensitivity and specificity of the frozen sections were 100% (95% CI: 87.7% − 100%) and 75% (95% CI: 19.4% − 99.4%), respectively. The AUC of the DiveScope and the frozen sections were 0.946 and 0.875 respectively, with no statistical difference(P = 0.578). Concerning time consumption, in tumorous samples with ≥ 0.8 cm in diameter, the Divescope cost 10.7 minutes on average while frozen pathology cost 41.9 minutes (p < 0.0001). For tumor samples < 0.8 cm in diameter, DiveScope cost an average of 10.7 minutes while the frozen pathology cost 50.1 minutes (p < 0.01). Compared with traditional pathologic examination of frozen sections, DiveScope was faster and displayed an equal accuracy for judging tumor margins intraoperatively.

DiveScope

Glioma

Histopathology

Sensitivity

Specificity

Surgical margins

Central nervous system (CNS) tumors, which can be categorized as primary or secondary tumors depending on the location of the primary site, are rarely seen compared to other types of tumors.[1, 2] Primary CNS tumors occur in approximately 30 out of every 100,000 people in the United States.[3, 4] Surgical resection is the first choice for most patients with primary CNS tumors, and its extent is one of the most decisive factors affecting the prognosis of CNS tumors, especially in glioblastoma. [5]

Currently, the mainstream surgical principle for glioma is maximal surgical resection consistent with the preservation of neurologic function,[6–9] as gross total resection (GTR) has the potential to prolong the GB patients’ overall survival by 2–3 months compared with subtotal resection.[10] However, due to the limits of the current surgical facilities and the techniques or the location of tumors, it is challenging to achieve GTR for all GB patients, and this might increase the recurrence rate.[10]

How to accurately and rapidly delineate glioma edge has been an area of exploration for years.[11–15] Nevertheless, pathology must be involved for the final judgment of whether the surgeon resects the tumor completely or not. Although frozen pathology can aid in the identification of tumor margins, its lagged effect leads to inconvenience in real-time recognition during surgery.[16, 17] In dermatology, there is an extremely efficient method in resecting skin cancer known as Mohs micrographic surgery because the margins of the removed tissue are inspected much more thoroughly under the microscope than with regular surgery.[18] The surgeon acted as both surgeon and pathologist at the same time during surgery.[19] Inspired by Mohs micrographic surgery, DiveScope, a handheld microscope, was developed for surgeons to observe resected tissue on a cellular level in real time.[20]

The DiveScope is a novel miniaturized histopathologic microscope developed by Max Act Precision Instruments (Suzhou) Ltd. that combines the collection and processing of histopathologic images, and it displays an image in real-time, which is not labeled for use temporarily. First, a light-emitting diode (LED) light source is connected to a microscopic imaging amplification system that emits blue light into the tissue that is observed through light-transmitting bundles. Tissue that is fluoresced in advance absorbs blue light and gives out green light that is received back by the microscopic imaging amplification system to amplify the signals. Next, the camera system collects and processes the magnified images, after which they are eventually transmitted and presented on the display. It can magnify tissue, such as the brain, mammary glands, or cervix, with a superb resolution to visualize sub-micron organelles such as nuclei, nucleoli, and cytosol. In head and neck cancer, it has shown high accuracy in identifying the benignity/malignancy of the tissue and the positivity/negativity of resection margins, which were highly consistent with those of intraoperative frozen section pathology tests.[20]

The purpose of our work was to compare the results of the DiveScope as an intraoperative histopathologic microscope for identifying surgical margins with frozen pathology section reports to assess the diagnostic value of the DiveScope. In addition, we compared the time consumption difference between the DiveScope and frozen sections for initial judgment. Furthermore, we also introduced a typical case of glioma for elucidation of the clinical application of the DiveScope in neurosurgery.

Animal experiments

The protocol for the animal experiments in this study was approved by the institutional review board of Gateway Medical Innovation Center (“Gateway” for short) (IACUC: SH2022-03004). Moreover, all experimental pigs were also purchased from the Gateway. To empty the stomach contents of the experimental pig, fasting was required 12 hours before surgery. Preoperative pig anesthesia and preoperative surgical area preparation were performed according to standard Gateway operating protocols. After induction of anesthesia, the animals were weighed, antibiotics were administered prophylactically, and an indwelling needle was placed into the ear vein for rehydration of intravenous fluids.

The experimental pig was placed in the prone position with the head fixed in the middle. The incision area was routinely disinfected before clothing. A single incision was made in the scalp, subcutaneous tissue, galea aponeurosis, and temporal muscle after the temporal muscle was rotated anteriorly. A bur was used to expose the dura mater and the dura was cut to expose brain tissue later. After excavating brain tissue of 1cm³-3cm³, we carefully checked the operation cavity for no active bleeding, and the DiveScope system was used to observe the brain tissue in the cavity in vivo after staining directly in the cranial cavity. 20 images were collected using the DiveScope system. In the end, we closed the cranial cavity layer by layer.

Patient Selection

Our prospective ex-vivo cohort study enrolled 40 consecutive patients from November 2021 to May 2022 in the Department of Neurosurgery at the First Affiliated Hospital of Sun Yat-sen University. Written consent is obtained from all patients according to our hospital-approved protocol. Inclusion criteria consisted of glioma, including grade I-IV. Exclusion criteria: lack of either frozen sections, DS test, or permanent paraffin pathology for the same site. Finally, 32 samples from 18 patients were recruited. Patients ranged in age from 17 to 73 years. More detailed patient demographics see in Table 1.

Table 1

Baseline characteristics of the 18 patients
Baseline characteristics
Age (Years) (mean, SD)	46.6 (12.8)	Preoperative MRI diagnosis
Gender		High-grade glioma	7
Male	12	Low-grade glioma	8
Female	6	Meningioma	1
Left/right		Ependymoma	1
Left	13	Dysembryoplastic neuroepithelial tumor, DNET	1
Right	4
Midline	1
Location		Peripheral edema
Frontal lobe	6	Yes	8
Temporal lobe	7	No	10
Basal ganglia	2	Pathologic reports
Parieto-occipital lobe	2	High-grade glioma
The fourth ventricle	1	WHO grade 3	2
		WHO grade 4	8
Degree of resection		Low-grade glioma
Gross total resection	15	WHO grade 1	2
Subtotal resection	3	WHO grade 2	6
MRI = Magnetic Resonance Imaging.
DNET = Dysembryoplastic neuroepithelial tumor.

Specimen preparation

We collected a portion of the margin of the resected fresh samples to stain immediately as the samples were obtained from the operating table. Before staining, the tissue was cut into two pieces, which separately were kept for visualization by the DiveScope or sent to the department of pathology for the frozen sections. The pieces of tissue would eventually be sent to the pathology department later for hematoxylin and eosin (HE) staining for definitive pathologic diagnosis. (FIG. 1)

For the staining procedure, we first mixed 0.5 ml standard fluorescein sodium (3 ml: 0.6 g) with 49.5 ml 0.9% sodium chloride thoroughly. Next, 2 ml of the mixture was extracted and placed in a container after which it was diluted with 0.9% sodium chloride until it became a solution of 0.2% sodium fluorescein. We then stained the sample with the diluted sodium fluorescein solution for 30 sec, the sample was then washed with 0.9% sodium chloride 3 times. Subsequently, the sample was infiltrated with a 1% methylene blue solution for 90 sec and then washed with 0.9% sodium chloride three times prior to observation under DiveScope.

Diagnosis of the pathologic images

The diagnosis process is double-blind. Prior to the trial, two surgeons were trained with 30 DiveScope’s images of glioma margins. The images were provided by Max Act Precision Instruments. Then two experienced neurosurgeons in the operating room were required to cooperate to make judgments about the margins of the removed tissue according to the DiveScope figures. If there appeared to be disagreement regarding any image, a final decision was made following a discussion between them. The 2 neurosurgeons interpreting the DiveScope images are blinded to the gross (surgeon's) impression of residual tumor vs no residual tumor. The judge criteria of margin status is based on the total area of heterogeneous nuclei whether over 1/50 in the visual field. Any less than 1/50 means the heterotypic cells are not obvious and are labeled negative margin, while any greater than 1/50 means positive. Atypic cells are typically characterized by increased density, enhanced nuclear staining, irregular nuclear shapes, and/or enlarged nuclei. On the other hand, one pathologist with at least 10 years of experience with tumor pathologic diagnosis was responsible for frozen section interpretation. Meanwhile, we will time the period from the start of staining until the initial report is issued for both diagnostic methods.

Statistical analysis

All statistical analyses were conducted by IBM SPSS Statistics Version 26.0 and the Medcalc statistical software package V19.0.4. The data analysis in our work includes three parts. In part one, diagnostic accuracy was compared between DiveScope and frozen modalities using 2-sample z tests. In this part, postoperative routine pathology results of patients were used as the gold standard. AUC, Sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio were calculated when possible with 95% CI. When P values below 0.05 are considered statistical differences.

In part two, kappa values were calculated to analyze the consistency of the DiveScope, frozen section, and permanent pathological paraffin result. No correlations were considered when the P-value is less than 0.05.

In part three, the comparison of the time consumption difference between the DiveScope and intraoperative frozen section pathological examinations for the initial judgment was assessed. When P values below 0.05 are considered statistical differences.

The structure of the final product of the DiveScope with a superb resolution to visualize sub-micron organelles

In the final version of the DiveScope, the surgery microscope collection, and processing system consists of the mainframe of an imaging system, a camera readout, a microscope, a distal hood, a power cord, a monitor, and an LED light source (Fig. 2A). The optical components of the camera readout include a Complementary Metal-Oxide-Semiconductor (CMOS) camera, a C-Mount, a fluorescence cube, a relay lens, and a miniature objective. The camera system is generally composed of an image-processing host and a camera. The working principle of the camera system is to connect the lens body through the optical interface, convert the optical signals collected by the lens body into digital signals, input them to the image processing host, and then output them to the display after being processed by the image processing host. The image processing host is the control center of the whole camera system that is used to process surgical videos and images captured from the camera. In our device, a cold light bulb is a bulb that uses a similar light source principle to an LED. The working principle of the cold light source is that the power module supplies power to each functional module, and the program is integrated into the control board module. The control panel module includes three parts: the Medium Attachment Unit (MAU) module, the LED driver module, and the dimming module.

With superb resolution, the DiveScope visualized the resolution test chart and sub-micron organelles, such as nuclei, nucleoli, and cytosol, in vitro imaging (FIG. 2B, C).

Performance of the DiveScope in experimental animals

To preliminarily assess the performance of the DiveScope, we dissected 3 experimental pigs’ brains and observed them under the DiveScope (Fig. 3A). The outline of the neurons’ nuclei in the gray matter can be identified (Fig. 3B), which is much larger compared to the nuclei of glial cells in the white matter (Fig. 3C). These images showed the excellent performance of the DiveScope in distinguishing neurons from the gliocytes. After the surgery, we did not observe evident side effects on all experimental pigs.

Performance of the handheld microscope in clinical application

The patient demographic and clinical variables were abstracted via chart reviews (Table 1). In all the patients, the fluorescent tissues at the margin of the resection cavity were all detected by the DiveScope after the completed microscopic resection. In normal brain tissue, all the neurons are in a sparse and regular cell distribution, with only one large and round nucleus plus a long axon and multiple dendrites (Fig. 4A). In tumorous tissue, no matter what pathologic type it is, tumor cells have been found to have the common features of high density, disordered distribution, a low degree of differentiation, and variation in size. Other than patients included in the statistics who were confirmed to be glioma patients, we also observed DiveScope pictures of one metastatic tumor in order to preliminarily study the performance of DS in other diseases (Fig. 4B). In a patient with an initial diagnosis of “cerebral infarction” from the emergency room, we found a similar manifestation under DiveScope. Such a manifestation can easily be misdiagnosed as a brain tumor. It was considered to be a cerebral abscess with merely large lymphocyte and neutrophil infiltration according to the postoperative pathology (Fig. 4B).

Diagnostic accuracy of the DiveScope in resection margins

First, the blinded pathologist analysis of the hematoxylin and eosin (HE)-stained research biopsies, which were derived from the same tissue observed under the DiveScope previously, served as the gold standard for final comparison with the results observed through the DiveScope. As was shown in Table 2, the consistency of the DiveScope and HE-staining pathology is acceptable, with only 3 cases of FN and the Kappa value of 0.68. Meanwhile, the 2×2 contingency table of the frozen section pathology and the HE-staining pathology was also listed, with the Kappa value of 0.84 (Table 3).

Table 2

Performance of DiveScope to differentiate tumorous from normal tissue in 32 specimens
DiveScope	HE-staining Pathologic Results		Total
DiveScope	Tumorous Tissue	Normal Tissue	Total
Positive	25	0	25
Negative	3	4	7
Total	28	4	32
HE = hematoxylin and eosin.

Table 3

Performance of the frozen sections to differentiate tumorous from normal tissue in 32 specimens
Frozen Sections	HE-staining Pathologic Results		Total
Frozen Sections	Tumorous Tissue	Normal Tissue	Total
Positive	25	0	25
Negative	4	3	7
Total	29	3	32
HE = hematoxylin and eosin.

Next, we directly compared the interpretation from the DiveScope and the frozen sections. The diagnostic measures of accuracy were calculated to evaluate the ability of the DiveScope as a novel intraoperative miniaturized histopathologic device (Table 4). As a result, 32 samples collected from 18 patients were classified as tumorous or normal tissue by the DiveScope. Regarding HE-stained sections as the gold standard, the sensitivity and the specificity of the DiveScope were 88.29% (95% CI: 70.8% − 97.6%) and 100% (95% CI: 39.8% − 100.0%), respectively, implying its strong ability to identify the positive margin and rule out the negative margin. In contrast, the sensitivity and specificity of the frozen sections were 100% (95% CI: 87.7% − 100%) and 75% (95% CI: 19.4% − 99.4%), respectively. Moreover, the positive likelihood ratio (PLR) of the DiveScope was imponderable because of no false positive case in DiveScope, while in contrast, the negative likelihood ratio (NLR) was 0.11 (95% CI: 0.037–0.31). In the aspect of the ROC curve, the AUC of the DiveScope and the frozen sections were 0.946 and 0.875, respectively (Fig. 5). These results proved the high diagnostic value of the DiveScope as a novel intraoperative miniaturized histopathologic device compared to conventional frozen sections.

Table 4

Comparison between DiveScope and frozen sections to differentiate tumorous from normal tissue in 32 specimens
DiveScope	Frozen Section Pathologic Results		Total
DiveScope	Tumorous Tissue	Normal Tissue	Total
Positive	25	0	25
Negative	4	3	7
Total	29	3	32

Time consumption

To prove the superiority of the DiveScope in rapid intraoperative pathologic diagnosis, we divided the samples into two groups according to the median diameter of all the tumorous samples (0.8 cm) and analyzed the time consumption for DiveScope or frozen sections, respectively (Fig. 6A, B). In tumorous samples with ≥ 0.8 cm in diameter, Divescope cost 10.7 minutes on average while frozen pathology cost 41.9 minutes (p < 0.0001). In tumorous samples with < 0.8 cm in diameter, DiveScope cost 10.7 minutes on average while frozen pathology cost 50.1 minutes (p < 0.01). In short, the time consumption for tissue identification for the DiveScope is significantly much faster than for the frozen sections no matter what size the sample is.

A typical clinical case

In this section, we report a typical operation that utilized the DiveScope (Fig. 7). A 49-year-old man presented with severe headache and nausea without vomiting or dizziness for two weeks. Magnetic resonance imaging (MRI) revealed a mass in the right frontal lobe which was considered to be high-grade glioma with intratumoral hemorrhage. During the surgery, we utilized the anterior, posterior, interior, and lateral borders of the tumorous tissue for observation with the DiveScope and frozen sections at the same time (Fig. 7A). After gross tumor resection with the help of intraoperative frozen sections and the DiveScope (Fig. 7B), we did not find a residual contrast-enhancing tumor on the early postoperative MRI (Fig. 7C), and this preliminarily proved the effectiveness of the DiveScope for the clinical determination of surgical margins.

Rapidly and accurately identifying glioma margins is a challenge in glioma surgery[21]. For instance, preoperative investigations could utilize a large amount of information from contrast-enhanced magnetic resonance imaging (MRI)[22, 23], functional MRI[15, 24], contrast-enhanced CT, positron emission tomography (PET)[25] or magnetoencephalography (MEG)[14] to determine the optimal surgical approach and assess the possibility of complete resection. However, in practical applications, surgeons often found preoperative investigations can not reflect the real situation in surgery perfectly. Therefore, surgeons came up with intraoperative investigations to compensate for the shortcomings of preoperative investigations. Neuronavigation, allowing us to locate the tumor margin in real time with preoperative imaging, is becoming widely applied during neurosurgeries[22], while near the end of surgery, the shift of tumor volume and cerebrospinal fluid drainage might influence on the inaccuracy of intraoperative neuronavigation[13]. Hence, intraoperative MRI could help correct for the brain shift due to tumor resection[26]. 5-aminolevulinic acid (5-ALA), one of those fluorescent agents which selectively stained glioma cells, is another option for intraoperative identification of tumor margins[12], but it requires oral administration four hours before the surgery and could have skin photosensitivity as side effects[12, 27, 28]. Thus, 5-ALA is not approved by some countries and regions yet. Here, we presented preliminary evidence of the efficacy of real-time visualization of brain tumors using the prototype DiveScope fluorescence imaging device for clinical applications.

Our animal experiments demonstrate that DiveScope is safe and feasible to detect in vivo. Owing to ultra-high images in the anti-shake design in real-time, it can continuously provide tissue and cellular structures with subtle nucleoli and nuclei. This is critical for DiveScope to quickly distinguish tumor tissue from surrounding normal parenchyma during surgery. At the same time, the tissue and cell structure of DiveScope are highly similar to conventional paraffin pathology. Consequently, it is pretty practical for the interpretation standard directly follows HE histomorphology.

In human in vitro experiments, we used a typical case to show that clean intraoperative margins were detected by DiveScope as well as no residual tumor was identified by postoperative imaging, which further verified the feasibility of DiveScope in detecting glioma margins in humans.

Finally, and most importantly, we simultaneously compared the accuracy, consistency, and timeliness between the DiveScope and frozen Interpretation. The results indicate that the interpretation accuracy of DiveScope was superior to that of intraoperative frozen, although there was no statistical difference. The fundamental reason for this discrepancy is that intraoperative frozen is easily disrupted by ice crystals and tissue extrusion during sampling[29], which has a significant impact on frozen judgment. In contrast, DiveScope is non-invasive in real-time and can circumvents these limitations. Subsequent comparisons between DiveScope and frozen accuracy are likely to become larger as we interpretation experience accumulates.

In terms of time effectiveness, DiveScope is much higher than frozen reports. We performed group comparisons based on the tissue diameter of 0.8cm and found that DiveScope takes approximately 10.7 minutes to issue results in both small (< 0.8cm) and large (≥ 0.8cm) tissues. Whereas, the time of frozen reports increases as tissue size increases, with an increase in the median time from 41.9 minutes to 50.1 minutes. Obviously, frozen will greatly prolong the operation time, increase the surgical trauma, and be inconvenient for promotion. In the consistency test, frozen was found to correlate moderately with DiveScope, with a Kappa value of 0.54 (P = 0.578). Suggesting that DiveScope images could be provided to pathologists as an important reference for issuing intraoperative frozen reports in the future, thereby may substantially improving the accuracy of frozen diagnosis.

To date, there have been similar products entering the market, but DiveScope still has incomparable advantages compared to them. For instance, ZEISS CONVIVO® System, as a probe-based confocal laser endomicroscopy (pCLE), can provide a 500µm visual field with a resolution of up to 0.5µm, which is the same as DiveScope, but its image data is collected at user-defined scan rate at a maximum of 4 frames/s (FPS) while the scan rate of DiveScope can reach up to 60 FPS.[30] Although OLYMPUS EndoCyto can also reach up to 60 FPS at a scan rate and offer a 500µm visual field, its low resolution of 3µm becomes one of its shortcomings. Compared to Mauna Kea CellVizio, another pCLE, the DiveScope is also ahead in the field of the resolution rate (0.5µm versus 1µm), visual field (500µm versus 240µm), scan rate (60 FPS versus 12 FPS), etc.[31] In addition, although Raman spectroscopy has shown promising application in the accurate diagnosis of cancerous tissues[32, 33], due to the infrequency of Raman scattering events, pixel-by-pixel measurement of a surgical field of view is unfeasible as this quickly leads to imaging times of multiple hours for specimens as small as 1 mm².[34, 35] Even if they improved the protocol, such approaches still require 12 to 24 min for imaging.[36, 37] In aspects of anti-vibration, the DiveScope adds OIS technique to ensure the surgeon could voluntarily move the microscope without blurring the images, which does not appear on the products mentioned above.

Because the fluorescence staining of the sample is accomplished in vitro, it is safe and atoxigenic for all patients. However, the small amount of tissue we stained for observation under the DiveScope was not exactly the same tissue sent to the department of pathology for frozen sections, but the tissues were extremely close to each other. Hence, there might have been a deviation between the results of the pathological reports and our observations from the DiveScope. It is confirmed that staining tissue with fluorescein sodium and methylene blue would not affect the results of the subsequent routine pathology examination. In the future, we can use the same tissue to process both the fluorescence imaging and the pathologic observations, and this would be more supportive of our intraoperative observations. For a more precise determination of the surgical margins intraoperatively, direct fluorescence staining and observations in vivo during surgery would be the next research trend, and this would require follow-up exploration regarding the safety and effective concentration of fluorescein sodium and methylene blue solutions for patients with brain tumors. Furthermore, the current image collection was done in 2D, and further work should also incorporate 3D modeling of fluorescence staining in the tissue.

Another limitation of this work is its small sample size. Because our study took place in a general hospital where glioma patients account for merely a small fraction of the department's patients, the patient information we can collect is limited in such a short time. A small sample size might influence our conclusions since most tissues in our dataset are positive margins which could cause bias on the Kappa value. In the following work, we will continue collecting glioma patients’ samples for more data to enhance our evidence. In addition, we found the error bars on the frozen section histograms (Fig. 6) are quite large. This might partly be due to different workers who sent the samples to the department of pathology at each time, which is difficult to control.

In conclusion, the handheld microscope is an emerging technique with significant potential to improve clinical outcomes in patients with brain tumors. Compared with traditional intraoperative pathologic examination of frozen sections, DiveScope was faster and produced similar results to a traditional pathological examination regarding the accuracy of intraoperative margin assessments. Our approach produced a real-time color image of the resected tissue with micron resolution, simplifying the image interpretation for the operator. This facilitates real-time decision-making by guiding surgeons to potential positive margins in minutes and allowing operators to adjust the surgical plan in real-time.

Ethics approval and consent to participate

The protocol for the animal experiments in this study was approved by the institutional review board of Gateway Medical Innovation Center (“Gateway” for short) (IACUC: SH2022-03004). All the human tissues were obtained from the Department of Neurosurgery at the First Affiliated Hospital of Sun Yat-sen University. Tissues were obtained after patients’ written consent under a protocol approved by the institutional review board of the First Affiliated Hospital of Sun Yat-sen University (No. [2020]322). Written consent is obtained from all patients according to our hospital-approved protocol.

Consent for publication

Not applicable.

Availability of data and materials

All data generated or analysed during this study are included in this published article and its supplementary information files.

Competing interests

The authors declare that they have no competing interests.

Funding

This work was supported by the National Natural Science Foundation of China 82001307 (to Baoshu Xie) and 81802484 (to Kejun He), and the Natural Science Foundation of Guangdong Province 2018A030313549 (to Kejun He) and Guangdong Basic and Applied Basic Research Foundation 2019A1515110460 (to Baoshu Xie).

Authors' contributions

WW, BX, NZ and KH conceived and designed research; NZ led the project; WW, BX, XZ, CZ and TH performed experiments; WW, HS and XL performed data analysis and wrote the paper. All authors read and approved the final manuscript.

Acknowledgements

We thank all the contributors to these clinical studies. We would like to express our deep gratitude to all principal investigators at the sites where the studies were conducted. We also thank Chengzhou He for his writing and editorial assistance.

Authors’ information

¹Department of Neurosurgery, Institute of Precision Medicine, the First Affiliated Hospital of Sun Yat-sen University, Guangdong Provincial Key Laboratory of Brain Function and Disease, Sun Yat-sen University, Guangzhou, Guangdong 510080, China; and ²Yale School of Medicine, New Haven, Connecticut, USA

Kurokawa R, Kurokawa M, Baba A, Ota Y, Pinarbasi E, Camelo-Piragua S, Capizzano AA, Liao E, Srinivasan A, Moritani T. Major Changes in 2021 World Health Organization Classification of Central Nervous System Tumors.Radiographics: a review publication of the Radiological Society of North America, Inc2022:210236.
Louis DN, Perry A, Wesseling P, Brat DJ, Cree IA, Figarella-Branger D, Hawkins C, Ng HK, Pfister SM, Reifenberger G, et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neurooncology. 2021;23(8):1231–51.
Ostrom QT, Gittleman H, Truitt G, Boscia A, Kruchko C, Barnholtz-Sloan JS. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2011–2015. Neuro-oncology 2018, 20(suppl_4):iv1-iv86.
Komlodi-Pasztor E, Gilbert MR, Armstrong TS. Diagnosis and Management of Stroke in Adults with Primary Brain Tumor.Current oncology reports2022.
Mohamadian M, Ahmadi SS, Bahrami A, Ferns GA. Review on the Therapeutic Potential of Curcumin and its Derivatives on Glioma Biology.Neurochemical research2022.
Raco A, Esposito V, Lenzi J, Piccirilli M, Delfini R, Cantore G. Long-term follow-up of intramedullary spinal cord tumors: a series of 202 cases. Neurosurgery. 2005;56(5):972–81. discussion 972–981.
Ottenhausen M, Krieg SM, Meyer B, Ringel F. Functional preoperative and intraoperative mapping and monitoring: increasing safety and efficacy in glioma surgery. NeuroSurg Focus. 2015;38(1):E3.
Rodríguez-Camacho A, Flores-Vázquez JG, Moscardini-Martelli J, Torres-Ríos JA, Olmos-Guzmán A, Ortiz-Arce CS, Cid-Sánchez DR, Pérez SR et al. Macías-González MDS, Hernández-Sánchez LC : Glioblastoma Treatment: State-of-the-Art and Future Perspectives. International journal of molecular sciences 2022, 23(13).
Mansouri A, Lai C, Scales D, Pirouzmand F. A phase II pilot randomized controlled trial to assess the feasibility of the "supra-marginal" surgical resection of malignant glioma (G-SUMIT: Glioma supra marginal incision trial) study protocol. Pilot and feasibility studies. 2022;8(1):138.
Noorbakhsh A, Tang JA, Marcus LP, McCutcheon B, Gonda DD, Schallhorn CS, Talamini MA, Chang DC, Carter BS, Chen CC. Gross-total resection outcomes in an elderly population with glioblastoma: a SEER-based analysis. J Neurosurg. 2014;120(1):31–9.
Orosco RK, Tapia VJ, Califano JA, Clary B, Cohen EEW, Kane C, Lippman SM, Messer K, Molinolo A, Murphy JD, et al. Positive Surgical Margins in the 10 Most Common Solid Cancers. Sci Rep. 2018;8(1):5686.
Maugeri R, Villa A, Pino M, Imperato A, Giammalva GR, Costantino G, Graziano F, Gulì C, Meli F, Francaviglia N et al. With a Little Help from My Friends: The Role of Intraoperative Fluorescent Dyes in the Surgical Management of High-Grade Gliomas.Brain sciences2018, 8(2).
Senft C, Bink A, Franz K, Vatter H, Gasser T, Seifert V. Intraoperative MRI guidance and extent of resection in glioma surgery: a randomised, controlled trial. Lancet Oncol. 2011;12(11):997–1003.
Tarapore PE, Tate MC, Findlay AM, Honma SM, Mizuiri D, Berger MS, Nagarajan SS. Preoperative multimodal motor mapping: a comparison of magnetoencephalography imaging, navigated transcranial magnetic stimulation, and direct cortical stimulation. J Neurosurg. 2012;117(2):354–62.
Trinh VT, Fahim DK, Maldaun MV, Shah K, McCutcheon IE, Rao G, Lang F, Weinberg J, Sawaya R, Suki D, et al. Impact of preoperative functional magnetic resonance imaging during awake craniotomy procedures for intraoperative guidance and complication avoidance. Stereotact Funct Neurosurg. 2014;92(5):315–22.
Xiao A, Shen B, Shi X, Zhang Z, Zhang Z, Tian J, Ji N, Hu Z. Intraoperative Glioma Grading Using Neural Architecture Search and Multi-modal Imaging.IEEE transactions on medical imaging2022, Pp.
Jain S, Kaushal M, Choudhary A, Bhardwaj M. Comparative evaluation of squash smear and frozen section in the intraoperative diagnosis of central nervous system tumours. Cytopathol. 2022;33(1):107–13.
Etzkorn JR, Alam M. What Is Mohs Surgery? JAMA dermatology. 2020;156(6):716.
Mansouri B, Bicknell LM, Hill D, Walker GD, Fiala K, Housewright C. Mohs Micrographic Surgery for the Management of Cutaneous Malignancies. Facial Plast Surg Clin North Am. 2017;25(3):291–301.
Zhang Y, Xie M, Xue R, Tang Q, Zhu X, Wang J, Yang H, Ma C. A Novel Cell Morphology Analyzer Application in Head and Neck Cancer. Int J Gen Med. 2021;14:9307–14.
Leclerc P, Ray C, Mahieu-Williame L, Alston L, Frindel C, Brevet PF, Meyronet D, Guyotat J, Montcel B, Rousseau D. Machine learning-based prediction of glioma margin from 5-ALA induced PpIX fluorescence spectroscopy. Sci Rep. 2020;10(1):1462.
Krivosheya D, Prabhu SS, Weinberg JS, Sawaya R. Technical principles in glioma surgery and preoperative considerations. J Neurooncol. 2016;130(2):243–52.
Fouke SJ, Benzinger T, Gibson D, Ryken TC, Kalkanis SN, Olson JJ. The role of imaging in the management of adults with diffuse low grade glioma: A systematic review and evidence-based clinical practice guideline. Journal of neuro-oncology 2015, 125(3):457–479.
Kuchcinski G, Mellerio C, Pallud J, Dezamis E, Turc G, Rigaux-Viodé O, Malherbe C, Roca P, Leclerc X, Varlet P, et al. Three-tesla functional MR language mapping: comparison with direct cortical stimulation in gliomas. Neurology. 2015;84(6):560–8.
Jansen NL, Suchorska B, Wenter V, Eigenbrod S, Schmid-Tannwald C, Zwergal A, Niyazi M, Drexler M, Bartenstein P, Schnell O, et al. Dynamic 18F-FET PET in newly diagnosed astrocytic low-grade glioma identifies high-risk patients. J nuclear medicine: official publication Soc Nuclear Med. 2014;55(2):198–203.
Knauth M, Wirtz CR, Tronnier VM, Aras N, Kunze S, Sartor K. Intraoperative MR imaging increases the extent of tumor resection in patients with high-grade gliomas. AJNR Am J Neuroradiol. 1999;20(9):1642–6.
Schebesch KM, Proescholdt M, Höhne J, Hohenberger C, Hansen E, Riemenschneider MJ, Ullrich W, Doenitz C, Schlaier J, Lange M, et al. Sodium fluorescein-guided resection under the YELLOW 560 nm surgical microscope filter in malignant brain tumor surgery–a feasibility study. Acta Neurochir. 2013;155(4):693–9.
Teixidor P, Arráez M, Villalba G, Garcia R, Tardáguila M, González JJ, Rimbau J, Vidal X, Montané E. Safety and Efficacy of 5-Aminolevulinic Acid for High Grade Glioma in Usual Clinical Practice: A Prospective Cohort Study. PLoS ONE. 2016;11(2):e0149244.
Chand P, Amit S, Gupta R, Agarwal A. Errors, limitations, and pitfalls in the diagnosis of central and peripheral nervous system lesions in intraoperative cytology and frozen sections. J Cytol. 2016;33(2):93–7.
Acerbi F, Pollo B, De Laurentis C, Restelli F, Falco J, Vetrano IG, Broggi M, Schiariti M, Tramacere I, Ferroli P, et al. Ex Vivo Fluorescein-Assisted Confocal Laser Endomicroscopy (CONVIVO® System) in Patients With Glioblastoma: Results From a Prospective Study. Front Oncol. 2020;10:606574.
Mondal A, Lackey J, Saeed M, Wu FY, Johnson JK, Huang C, Sachse FB, Hitchcock R, Kaza AK. An Imaging Protocol to Discriminate Specialized Conduction Tissue During Congenital Heart Surgery. Semin Thorac Cardiovasc Surg. 2019;31(3):537–46.
Horgan CC, Bergholt MS, Thin MZ, Nagelkerke A, Kennedy R, Kalber TL, Stuckey DJ, Stevens MM. Image-guided Raman spectroscopy probe-tracking for tumor margin delineation.Journal of biomedical optics2021, 26(3).
Orringer DA, Pandian B, Niknafs YS, Hollon TC, Boyle J, Lewis S, Garrard M, Hervey-Jumper SL, Garton HJL, Maher CO et al. Rapid intraoperative histology of unprocessed surgical specimens via fibre-laser-based stimulated Raman scattering microscopy.Nature biomedical engineering2017,1.
Pence I, Mahadevan-Jansen A. Clinical instrumentation and applications of Raman spectroscopy. Chem Soc Rev. 2016;45(7):1958–79.
Hubbard TJE, Shore A, Stone N. Raman spectroscopy for rapid intra-operative margin analysis of surgically excised tumour specimens. Analyst. 2019;144(22):6479–96.
Kong K, Rowlands CJ, Varma S, Perkins W, Leach IH, Koloydenko AA, Williams HC, Notingher I. Diagnosis of tumors during tissue-conserving surgery with integrated autofluorescence and Raman scattering microscopy. Proc Natl Acad Sci USA. 2013;110(38):15189–94.
Shipp DW, Rakha EA, Koloydenko AA, Macmillan RD, Ellis IO, Notingher I. Intra-operative spectroscopic assessment of surgical margins during breast conserving surgery. Breast cancer research: BCR. 2018;20(1):69.

Download PDF

Version 1

posted

You are reading this latest preprint version

Application of a Novel Miniaturized Histopathologic Microscope for Identifying Cerebral Glioma Margins Rapidly During Surgery

Status:

Version 1

Abstract

Figures

Introduction

Methods

Animal experiments

Patient Selection

Specimen preparation

Diagnosis of the pathologic images

Statistical analysis

Results

Performance of the DiveScope in experimental animals

Performance of the handheld microscope in clinical application

Diagnostic accuracy of the DiveScope in resection margins

Time consumption

A typical clinical case

Discussion

Limitations

Declarations

References

Status:

Version 1