Successful application of dietary ketogenic metabolic therapy in patients with glioblastoma: a clinical study

FIGURE 1 (A) Simplified scheme of glucose and ketone metabolism in a normal brain cell.

• Overview of glucose and ketone metabolism: Figure 1A illustrates the metabolic processes within a normal brain cell, focusing on glucose and ketone metabolism. Glucose enters the cell via a glucose transporter (Glut-1) and is converted to pyruvate through glycolysis in the cytoplasm. Pyruvate then enters the mitochondria, where it's converted to Acetyl-CoA, initiating the Krebs cycle (also known as the citric acid cycle). This cycle produces NADH and ATP, which are essential for cellular energy. When glucose is limited, ketone bodies enter the cell via monocarboxylate transporters (MCTs) and are converted to Acetyl-CoA, feeding into the Krebs cycle to produce energy.
• Integration of metabolic pathways: The diagram depicts the interconnectedness of glucose and ketone metabolism in a normal brain cell. The figure shows that both glucose and ketone bodies are metabolized to Acetyl-CoA, which enters the Krebs cycle. The Krebs cycle is a series of chemical reactions that extract energy from Acetyl-CoA and generate molecules like NADH and ATP, which the cell uses for fuel.
• Role of transporters: The figure highlights the role of specific transporters, Glut-1 for glucose and MCTs for ketone bodies, in facilitating the entry of these substrates into the cell. These transporters are crucial for cellular metabolism because they regulate the availability of energy substrates inside the cell.

• Accuracy of metabolic representation: The diagram accurately represents the core biochemical pathways of glucose and ketone metabolism in a normal brain cell. However, it simplifies the complexity of these processes by omitting regulatory enzymes and intermediate steps, which could be misleading for experts seeking detailed mechanistic insights.
• Completeness of metabolic pathways: The schematic representation does not include all possible metabolic fates of pyruvate or Acetyl-CoA, such as lipid synthesis or amino acid metabolism, which are also relevant in brain cells. This omission could oversimplify the metabolic capabilities of normal brain cells.
• Inclusion of regulatory mechanisms: The figure does not explicitly address the regulation of glucose and ketone metabolism, such as hormonal control or allosteric regulation of key enzymes. Including these regulatory aspects would provide a more complete picture of metabolic control in normal brain cells.

• Overall clarity and visual representation: The diagram effectively uses visual cues to represent metabolic pathways, making it easier to understand the complex processes involved. The use of arrows and labels is clear, but could benefit from more specific annotations to highlight key regulatory steps or enzyme involvement.
• Caption completeness: The caption provides a basic description of the figure's content, but could be expanded to include the specific context within the study (i.e., how this normal cell metabolism contrasts with that of a cancer cell) to improve its immediate relevance to the reader.
• Accessibility for non-experts: For readers unfamiliar with metabolic pathways, a brief glossary or more detailed introduction to the key molecules and processes (e.g., glycolysis, Krebs cycle) would enhance understanding.

• Overview of ketogenic diet impact: Figure 1B illustrates how a ketogenic diet affects metabolism in a normal brain cell. A ketogenic diet is a diet very low in carbohydrates, moderate in protein, and high in fats. When carbohydrates are restricted, the body produces ketone bodies from fat, which serve as an alternative fuel source. In this diagram, the ketogenic diet is shown to inhibit glucose uptake by the brain cell, represented by a crossed-out glucose transporter (Glut-1).
• Shift in substrate utilization: The figure shows that ketone bodies enter the cell via monocarboxylate transporters (MCTs). Once inside, they're converted to Acetyl-CoA, which then enters the Krebs cycle to produce energy. The ketogenic diet forces the brain cell to rely more heavily on ketone bodies for fuel instead of glucose.
• Metabolic consequences: The diagram highlights the reduced reliance on glucose and increased utilization of ketone bodies in normal brain cells during a ketogenic diet. This metabolic shift is thought to have therapeutic benefits in certain neurological conditions.

• Accuracy of metabolic representation: The diagram accurately represents the core concept of reduced glucose utilization and increased ketone body utilization in normal brain cells under a ketogenic diet. However, it simplifies the complex regulatory mechanisms that govern substrate selection and metabolic flux.
• Completeness of adaptive mechanisms: The figure does not address the potential adaptive mechanisms that brain cells might employ to compensate for reduced glucose availability, such as increased expression of ketone body transporters or altered enzyme activity. This omission could limit the figure's scientific rigor.
• Inclusion of energy status: The figure does not explicitly show the effects on ATP production or energy status within the cell, which are key outcomes of the metabolic shift. Including this information would strengthen the figure's scientific validity.

• Visual representation of metabolic shift: The diagram effectively illustrates the shift in substrate utilization in normal brain cells under a ketogenic diet. The visual representation of reduced glucose entry and increased ketone body entry is clear. However, the figure could benefit from a more explicit depiction of the relative concentrations or flux rates of these substrates to better convey the quantitative impact of the diet.
• Contextual information in caption: The caption provides context for the figure, but it could be enhanced by mentioning the intended benefit or consequence of this metabolic shift in normal brain cells, such as neuroprotection or energy efficiency.
• Accessibility for non-experts: While the diagram is simplified, it assumes a certain level of familiarity with metabolic pathways. Providing a brief explanation of the rationale behind the ketogenic diet's impact on glucose and ketone body utilization would improve accessibility for readers without a strong background in metabolism.

• Overview of cancer cell metabolism: Figure 2A depicts the metabolic processes within a cancer cell, emphasizing the Warburg effect, which is the preference for glycolysis (the breakdown of glucose) over oxidative phosphorylation (the use of oxygen to generate energy in the mitochondria) even when oxygen is available. Glucose enters the cell and is converted to pyruvate, but instead of primarily entering the mitochondria, it's converted to lactate via anaerobic glycolysis.
• Mitochondrial dysfunction: The figure indicates that the mitochondria in cancer cells are dysfunctional. This is represented by dashed lines around the Krebs cycle and reduced ATP production, implying that oxidative phosphorylation is impaired. The cancer cell relies heavily on glycolysis for its energy needs.
• Lactate production and tumor microenvironment: The diagram shows the production of lactate, which is exported from the cell. Lactate production contributes to an acidic environment around the tumor, which can promote cancer progression and metastasis. Acidosis, lactate and MCTs are highlighted.

• Accuracy of metabolic representation: The diagram accurately represents the core features of cancer cell metabolism, including increased glycolysis and mitochondrial dysfunction. However, it is a simplified representation and does not capture the full complexity of metabolic heterogeneity within tumors.
• Completeness of metabolic pathways: The figure does not include other important metabolic pathways that are often dysregulated in cancer cells, such as glutamine metabolism or fatty acid synthesis. Including these pathways would provide a more complete picture of cancer cell metabolism.
• Inclusion of regulatory mechanisms: The figure does not explicitly address the genetic and epigenetic factors that contribute to the metabolic rewiring of cancer cells. Mentioning key oncogenes or tumor suppressor genes that regulate metabolism would enhance the figure's scientific validity.

• Clarity of visual representation: The diagram clearly illustrates the metabolic differences in cancer cells compared to normal cells, specifically highlighting the reliance on glycolysis and reduced mitochondrial function. The use of simplified schematics effectively conveys the overall metabolic phenotype.
• Caption completeness: The caption provides a basic description, but could be enhanced by explicitly stating the Warburg effect and how it is represented in the figure. Mentioning the therapeutic implications (i.e., targeting glycolysis) would also improve the caption's impact.
• Accessibility for non-experts: For readers unfamiliar with cancer metabolism, a brief explanation of why cancer cells favor glycolysis even in the presence of oxygen (Warburg effect) would improve understanding.

• Overview of ketogenic diet impact: Figure 2B illustrates the effects of a ketogenic diet on cancer cell metabolism. The diagram shows a crossed-out glucose transporter (Glut-1), indicating that the ketogenic diet reduces glucose uptake into the cancer cell. This is because the ketogenic diet aims to lower blood glucose levels, thus depriving cancer cells of their preferred fuel source.
• Impaired ketone body oxidation: The diagram indicates that even though ketone bodies are available, the cancer cell's ability to oxidize them efficiently is impaired due to mitochondrial abnormalities. This is represented by the dashed lines around the Krebs cycle, implying that the cancer cell cannot effectively use ketone bodies for energy production.
• Therapeutic consequences: The figure highlights that the combination of reduced glucose uptake and impaired ketone body oxidation leads to reduced proliferation rates in cancer cells. This is the intended therapeutic effect of the ketogenic diet.

• Accuracy of metabolic representation: The diagram accurately represents the intended metabolic effects of a ketogenic diet on cancer cells, including reduced glucose availability and impaired ketone body oxidation. However, it is a simplified representation and does not capture the full complexity of metabolic adaptations that cancer cells can undergo.
• Completeness of metabolic pathways: The figure does not address the potential for cancer cells to upregulate other metabolic pathways, such as glutamine metabolism or fatty acid oxidation, to compensate for the reduced glucose availability. Including these alternative pathways would provide a more complete picture of cancer cell metabolism.
• Inclusion of microenvironmental factors: The figure does not explicitly address the role of the tumor microenvironment or systemic factors that can influence cancer cell metabolism. Including these factors would enhance the figure's scientific validity.

• Clarity of visual representation: The diagram effectively illustrates the intended impact of a ketogenic diet on cancer cells, specifically highlighting the reduction in glucose availability and the potential for reduced proliferation rates. The use of crossed-out arrows effectively conveys the inhibition of glucose uptake.
• Caption completeness: The caption could be enhanced by explicitly mentioning the therapeutic rationale, which is to exploit the metabolic inflexibility of cancer cells and their dependence on glucose. Also, stating the potential limitations (i.e., cancer cells may still use ketone bodies) would improve the caption's completeness.
• Accessibility for non-experts: For readers unfamiliar with cancer metabolism, a brief explanation of why limiting glucose might be detrimental to cancer cell growth would improve understanding. It could also benefit from acknowledging the potential for metabolic adaptation.

• Overview of patient characteristics: Table 1 presents the characteristics of all 18 patients who participated in the study. The table includes the following information for each patient: patient identifier (P), gender (G), age (A), date of diagnosis (DD), date of surgery and type of operation (DS-TOO), chemotherapy and radiation therapy administered before the ketogenic diet (Chemo + Rad prior KD administration), and molecular biology (MB).
• Patient demographics and surgical information: The table shows a mix of male (M) and female (F) patients, with ages ranging from 34 to 75 years. The 'DS-TOO' column indicates whether the patient underwent total resection or subtotal resection/stereotactic biopsy. Total resection refers to complete removal of the tumor, while subtotal resection means that part of the tumor was left.
• Treatment and molecular biology information: The 'Chemo + Rad prior KD administration' column lists the specific chemotherapeutic agents (Temozolamide) and the use of radiation therapy, specifically mentioning 30 cycles of radiation. The 'MB' column indicates the IDH1 status of the tumor, with results showing either negative (-), positive (+), or IDH 1-2.

• Relevance of included variables: The table provides relevant descriptive information about the patient cohort, which is essential for assessing the study's generalizability. The inclusion of molecular biology data (IDH1 status) is important for characterizing the tumor subtypes.
• Completeness of clinical data: The table is missing key clinical data that could influence the outcomes, such as Karnofsky Performance Status (KPS) scores at baseline, ECOG scores, extent of resection, and details of concurrent treatments. Adding these variables would strengthen the table's scientific value.
• Transparency of data collection: The table does not indicate the source of the patient data or the methods used to collect it. Clearly stating the data collection procedures would improve the table's transparency and scientific rigor.

• Overall organization and presentation: The table is generally well-organized, providing essential information about the study participants. However, the table could benefit from including summary statistics (e.g., mean ± SD for age) to allow for easier comparison between groups.
• Clarity of abbreviations: The use of abbreviations is generally defined, but a glossary of abbreviations within the table or in the caption would ensure clarity for all readers.
• Relevance to study objectives: The table effectively presents the data to support the study's methods. However, consider adding a column indicating whether each patient adhered to the ketogenic diet and for how long, as this is a critical variable in the study.

• Overview of the ketogenic diet example: Table 2 provides an example of a ketogenic diet with a total daily calorie intake of 2,150 Kcal and a 2:1 ketogenic ratio. The ketogenic ratio refers to the ratio of fat to combined protein and carbohydrates in the diet. The table lists specific food items and their quantities for morning, snack, lunch, snack, and dinner meals.
• Specific food items for main meals: The morning meal consists of 17g tuna in oil and 8g olive oil. The lunch meal includes 93g sardines and 41g olive oil, while dinner consists of 117g raw green salad, 93g salmon, and 41g olive oil.
• Specific food items for snacks: The snack options include items like 20g feta cheese, 60g keto-focaccia, 1 egg fortified with w-3 fatty acids, and 15g avocado. The table provides a 'Daily menu plan' indicating the distribution of food items across the day.

• Lack of design criteria: The table provides an example of a ketogenic diet, but it lacks information on the specific criteria used to design the diet (e.g., target blood ketone levels, individual patient needs). Including this information would improve the table's scientific rigor.
• Missing source of food composition data: The table does not specify the source of the food composition data used to calculate the macronutrient content of the diet. Providing this information would enhance the table's transparency and scientific validity.
• Nutritional completeness: The table presents a single example, but it doesn't indicate whether this diet is nutritionally complete or if it was supplemented with vitamins or minerals. Including this information would improve the table's scientific validity.

• Completeness of macronutrient information: The table provides a useful example of a ketogenic diet, but it lacks information on the macronutrient breakdown (grams of fat, protein, carbohydrates) for each meal. Including this information would allow readers to assess the diet's composition and adherence to the 2:1 ratio more effectively.
• Inclusion of meal timing: The table presents a daily menu plan, but it does not specify the timing of the meals or snacks. Including this information would provide a more complete picture of the dietary regimen.
• Representativeness of the example: The table provides a specific example, but it doesn't indicate whether this was a standard diet or if it was individualized based on patient preferences or tolerances. Clarifying this aspect would enhance the table's value.

FIGURE 3 Patient 1: (A) Pre-operative brain MRI (T2/FLAIR) (B) Pre-operative brain MRI (T1 with contrast) (C) 38-month follow-up brain MRI (T1 with contrast) (D) 80-month follow-up brain MRI (T1 with contrast).

• Pre-operative MRI characteristics: Figure 3 presents a series of brain MRIs for Patient 1, a 41-year-old man diagnosed with Glioblastoma Multiforme (GBM). Panel A shows a pre-operative MRI using T2-weighted Fluid Attenuated Inversion Recovery (FLAIR) sequence, which is sensitive to fluid and edema. Panel B shows a pre-operative MRI using T1-weighted sequence with contrast enhancement, which highlights areas of blood-brain barrier disruption.
• Follow-up MRI characteristics and key observation: Panel C shows a 38-month follow-up MRI using T1-weighted sequence with contrast. Panel D shows an 80-month follow-up MRI using T1-weighted sequence with contrast. The key observation is the absence of any significant contrast enhancement or tumor recurrence in the follow-up scans, suggesting a stable condition.
• Long-term stability and treatment efficacy: The sequential MRIs provide a visual timeline of the patient's condition, demonstrating the long-term stability achieved with the treatment regimen, including the ketogenic diet. The absence of tumor recurrence over 80 months is a significant finding.

• Appropriateness of MRI sequences: The figure provides valuable visual evidence of the patient's response to treatment. The use of T2/FLAIR and T1 with contrast sequences is appropriate for assessing tumor characteristics and recurrence.
• Missing MRI acquisition parameters: The figure lacks information on the MRI acquisition parameters (e.g., field strength, slice thickness). Including this information would improve the figure's scientific rigor and allow for better comparison with other studies.
• Lack of quantitative measurements: The figure does not include quantitative measurements of tumor volume or contrast enhancement. Including these measurements would provide a more objective assessment of the treatment effect.
• Scale and orientation of images: Based on the reference text, the figure has multiple panels depicting the same patient over time, therefore, it is important that the scale and orientation of the images are maintained for easy comparison.

• Visual demonstration of treatment efficacy: The figure effectively demonstrates the long-term stability of the patient's condition following treatment and adherence to the ketogenic diet. The sequential MRIs clearly show the absence of tumor recurrence over a significant period.
• Caption conciseness and key takeaway: The caption is comprehensive in listing the MRI sequences and time points, but it could benefit from briefly stating the key observation (i.e., absence of tumor recurrence) to immediately draw the reader's attention.
• Accessibility for non-experts: While the figure clearly shows the absence of recurrence, it assumes the reader has some familiarity with MRI interpretation. Including a brief annotation highlighting key anatomical landmarks or potential signs of recurrence would improve accessibility for non-experts.

FIGURE 3 The patient reports a residual mild anomia and is currently working as a teacher.

• Description of patient's self-reported condition: The caption describes the patient's self-reported condition, noting a 'residual mild anomia,' which is a type of aphasia characterized by difficulty recalling names of objects, people, and places. The patient is also described as currently working as a teacher, which indicates a high level of cognitive and functional recovery.
• Indication of functional recovery: The caption highlights the patient's ability to maintain employment as a teacher, suggesting a significant improvement in their quality of life and functional abilities despite the presence of a neurological deficit.
• Complementary information to MRI findings: The information in the caption complements the MRI findings, providing a holistic view of the patient's response to treatment. While the MRI shows no tumor recurrence, the caption provides insight into the patient's cognitive function.

• Subjectivity of patient-reported information: The caption provides patient-reported information, which is subjective. While valuable for understanding the patient's experience, it should be interpreted in the context of objective measures of cognitive function.
• Need for objective cognitive assessment: The presence of 'residual mild anomia' suggests that formal neuropsychological testing might reveal other subtle cognitive deficits that are not apparent in the patient's daily activities. Conducting such testing and reporting the results would strengthen the study.
• Limitations of subjective assessment: The information is relevant to assessing the overall outcome of the treatment, but it is limited by the lack of standardized measures of cognitive function and quality of life. Including such measures would enhance the scientific rigor of the study.

• Contextual information on patient's functional status: The caption provides important context about the patient's functional status and quality of life. Stating that the patient is working as a teacher reinforces the positive outcome of the treatment.
• Balanced perspective and definition of medical terms: The mention of 'residual mild anomia' provides a balanced perspective, acknowledging that the patient is not entirely free of neurological deficits. Defining 'anomia' briefly (i.e., difficulty finding words) would further improve clarity for non-medical readers.
• Integration with MRI findings: Connecting this information directly to the MRI findings (i.e., 'despite a residual mild anomia, MRI shows no recurrence') would strengthen the link between the imaging data and the patient's clinical status.

FIGURE 4 Histopathology: Patient 1 (A) Typical Morphology of Glioblastoma (H&E x200).

• Overview of histopathological image: Figure 4, panel A, presents a histopathological image of a Glioblastoma Multiforme (GBM) sample from Patient 1. The image is stained with Hematoxylin and Eosin (H&E), a common staining technique used to visualize cell structures. The magnification is 200x, meaning the structures are magnified 200 times their actual size.
• Key cellular features revealed by H&E staining: The H&E staining allows for the visualization of key cellular features, such as cell shape, nuclear size and shape, and the presence of any abnormal structures. In GBM, typical features include high cellularity (many cells packed together), cellular pleomorphism (variation in cell size and shape), nuclear atypia (abnormal nuclear appearance), and the presence of mitotic figures (cells undergoing division).
• Confirmation of GBM diagnosis: The image confirms the diagnosis of GBM by demonstrating these characteristic histopathological features. The absence of specific numerical data (e.g., mitotic index, percentage of necrotic tissue) is typical for a qualitative histopathological assessment.

• Appropriateness of staining method: The image provides visual confirmation of the GBM diagnosis based on standard histopathological criteria. The use of H&E staining is appropriate for assessing cellular morphology and architecture.
• Missing diagnostic criteria: The figure lacks information on the specific criteria used to diagnose GBM (e.g., WHO criteria). Clearly stating the diagnostic criteria would improve the figure's scientific rigor.
• Representativeness of the image: The figure presents a single image, but it doesn't indicate whether this image is representative of the entire tumor sample. Including multiple images or a statement about representativeness would improve the figure's scientific validity.

• Visual confirmation of diagnosis: The image provides visual confirmation of the GBM diagnosis, which is essential for validating the study's inclusion criteria. However, the figure would benefit from including a scale bar to allow readers to assess the size and morphology of the cells more accurately.
• Caption completeness: The caption provides the staining method (H&E) and magnification (x200), which is important for interpreting the image. However, it could be enhanced by briefly describing the key features of GBM that are visible in the image (e.g., cellular pleomorphism, nuclear atypia).
• Accessibility for non-experts: For readers unfamiliar with histopathology, a brief explanation of what H&E staining reveals and what features are characteristic of GBM would improve understanding.

• Overview of endothelial hyperplasia: Figure 4, panel B, presents a histopathological image showing severe endothelial hyperplasia, a characteristic feature of Glioblastoma Multiforme (GBM). Endothelial hyperplasia refers to the abnormal proliferation of endothelial cells, which line the blood vessels. The magnification is 100x.
• Formation of glomeruloid bodies: In GBM, endothelial hyperplasia leads to the formation of glomeruloid bodies, which are clusters of abnormal blood vessels. These vessels are often leaky and contribute to the edema and angiogenesis that are characteristic of GBM.
• Severity of proliferation: The image demonstrates the severe proliferation of endothelial cells, which is a key diagnostic feature of GBM. The severity of the hyperplasia is a notable characteristic.

• Appropriateness of magnification: The image provides visual confirmation of endothelial hyperplasia, a well-established diagnostic criterion for GBM. The magnification (100x) is appropriate for visualizing this feature.
• Missing staining method: The figure lacks information on the staining method used. Clearly stating the staining method (e.g., H&E) would improve the figure's scientific rigor.
• Representativeness of the image: The figure presents a single image, but it doesn't indicate whether this image is representative of the area of most severe endothelial hyperplasia within the tumor sample. Including multiple images or a statement about representativeness would improve the figure's scientific validity.

• Visual highlighting of key feature: The image provides visual evidence of a key diagnostic feature of GBM: endothelial hyperplasia. However, the figure would benefit from including an arrow or annotation to specifically highlight the area of endothelial hyperplasia, as it might not be obvious to all readers.
• Caption completeness and explanation of key feature: The caption provides the magnification (X100), but it could be enhanced by briefly explaining what endothelial hyperplasia is (i.e., abnormal proliferation of blood vessel cells) and its significance in GBM.
• Accessibility for non-experts: For readers unfamiliar with histopathology, a brief explanation of what endothelial hyperplasia looks like under a microscope would improve understanding. Describing the morphology of the cells (e.g., multilayered, plump) would be helpful.

• Overview of ketogenic diet intake: Table 3 presents the total daily intake for the ketogenic diets used in the study. It includes data for multiple patients (Patient 1, Patient 2, Patient 3, Patient 4, Patient 5, and Patient 6) . The data includes Duration of the diet, Kcal (total calories per day), Fat g/d (grams of fat per day), Protein g/d (grams of protein per day), CHOs g/d (grams of carbohydrates per day), and KR (ketogenic ratio).
• Macronutrient breakdown for patients 1 and 2: For Patient 1, the table shows a Modified Ketogenic Diet with 2000 Kcal, 169 g/d fat, 100 g/d protein, 20 g/d CHOs, and a KR of 1.4:1. For Patient 2, the table shows a Mediterranean ketogenic diet with MCTS, with caloric intake from 2150 to 2298, fat from 189.2 to 213, protein from 73.9 to 75, CHOs from 19.6 to 20, and a KR above 2:1.
• Macronutrient breakdown for patients 3-6: The table also shows data for Patient 3 (Mediterranean Ketogenic Diet), Patient 4 (Mediterranean ketogenic diet with MCTs), Patient 5 (Mediterranean Ketogenic Diet), and Patient 6 (Mediterranean Ketogenic Diet), showing variations in caloric intake and macronutrient composition.

• Relevance of included variables: The table provides valuable information on the macronutrient composition of the ketogenic diets, which is essential for assessing the study's methodology. The inclusion of the ketogenic ratio is also important for characterizing the diets.
• Missing dietary assessment methods: The table does not specify the methods used to assess dietary intake (e.g., food diaries, 24-hour recalls). Clearly stating the methods used to collect dietary data would improve the table's scientific rigor.
• Lack of micronutrient information: The table does not include information on the micronutrient content of the diets. Including this information would improve the table's scientific validity, as micronutrient deficiencies can be a concern with ketogenic diets.

• Specificity of macronutrient sources: The table provides essential information on the macronutrient composition of the ketogenic diets used in the study. However, the table could be more informative by including the specific types of fats, proteins, and carbohydrates used in each diet. For example, specifying that the fats were primarily from olive oil and avocados, or that the proteins were from fish and chicken, would be helpful.
• Range of ketogenic ratios: The table includes data on daily calorie intake, fat, protein, and carbohydrate content, as well as the ketogenic ratio. However, it would be useful to include the range of ketogenic ratios achieved for each patient, rather than just providing a single example. This would give a better sense of the variability in dietary adherence.
• Individualization of diets: The table effectively presents the data, but it could benefit from a brief explanation of how these diets were individualized to meet the needs of each patient. For example, were any adjustments made based on activity level, weight, or other factors?

FIGURE 5 Patient 2: (A) pre-operative brain MRI (T2/FLAIR) (B) pre-operative brain MRI (T1 with contrast) (C) 20-month follow-up brain MRI (T1 with contrast) (D) 40-month follow-up brain MRI (T1 with contrast).

• Pre-operative MRI characteristics: Figure 5 presents a series of brain MRIs for Patient 2, a 59-year-old male. Panel A shows a pre-operative MRI using T2-weighted Fluid Attenuated Inversion Recovery (FLAIR) sequence, revealing a large heterogeneous mass in the right temporoparietal area. Panel B shows a pre-operative MRI using T1-weighted sequence with contrast enhancement, which highlights the tumor's vascularity and blood-brain barrier disruption.
• Follow-up MRI characteristics and key observation: Panel C shows a 20-month follow-up MRI using T1-weighted sequence with contrast. Panel D shows a 40-month follow-up MRI using T1-weighted sequence with contrast. The key observation is the change in size, shape and contrast enhancement after the ketogenic diet. The 53mm diameter from the reference text is not visible on the images themselves.
• Treatment response and long-term effects: The sequential MRIs provide a visual timeline of the patient's condition, demonstrating the response to treatment and the long-term effects of the ketogenic diet. The changes in the tumor's characteristics over time are visually apparent.

• Appropriateness of MRI sequences: The figure provides valuable visual evidence of the patient's response to treatment. The use of T2/FLAIR and T1 with contrast sequences is appropriate for assessing tumor characteristics and response to therapy.
• Missing MRI acquisition parameters: The figure lacks information on the MRI acquisition parameters (e.g., field strength, slice thickness). Including this information would improve the figure's scientific rigor and allow for better comparison with other studies.
• Lack of quantitative measurements: The figure does not include quantitative measurements of tumor volume or contrast enhancement. Including these measurements would provide a more objective assessment of the treatment effect.
• Scale and orientation of images: The scale and orientation of the images are maintained for easy comparison, and the figure is also easy to interpret.

• Visual demonstration of treatment response: The figure provides a clear visual timeline of the tumor's response to treatment in Patient 2. The use of pre- and post-operative MRIs is helpful for demonstrating the changes in tumor size and characteristics over time.
• Caption conciseness and key takeaway: The caption is comprehensive in listing the MRI sequences and time points, but it could benefit from briefly stating the key observations (e.g., tumor reduction, stable disease) to immediately draw the reader's attention.
• Accessibility for non-experts: While the figure clearly shows the changes in the tumor, it assumes the reader has some familiarity with MRI interpretation. Including brief annotations highlighting key features (e.g., tumor boundaries, areas of necrosis) would improve accessibility for non-experts.

FIGURE 6 Patient 3: (A) brain MRI on diagnosis (T2/FLAIR) (B) brain MRI on diagnosis (T1 with contrast) (C) 9-month follow up brain MRI (T1 with contrast) (D) GBM relapse, 12-month follow up brain MRI (T1 with contrast) (E) 30-month follow up brain MRI (T1 with contrast).

• MRI at diagnosis: Figure 6 presents a series of brain MRIs for Patient 3. Panel A shows the brain MRI at diagnosis using T2-weighted Fluid Attenuated Inversion Recovery (FLAIR) sequence, highlighting edema and lesions. Panel B shows the brain MRI at diagnosis using T1-weighted sequence with contrast enhancement, revealing the tumor's vascularity.
• Follow-up MRI characteristics: Panel C shows the 9-month follow-up brain MRI using T1-weighted sequence with contrast, demonstrating the initial response to treatment. Panel D shows GBM relapse at 12-month follow up using T1-weighted sequence with contrast, revealing the tumor re-growth. Panel E shows the 30-month follow up brain MRI using T1-weighted sequence with contrast.
• Visual timeline of tumor behavior: The sequential MRIs provide a visual timeline of the patient's condition, demonstrating the initial response to treatment and the subsequent relapse. The figure allows for a visual assessment of the tumor's behavior over time.

• Appropriateness of MRI sequences: The figure provides valuable visual evidence of the patient's response and subsequent relapse. The use of T2/FLAIR and T1 with contrast sequences is appropriate for assessing tumor characteristics and response to therapy.
• Missing MRI acquisition parameters: The figure lacks information on the MRI acquisition parameters (e.g., field strength, slice thickness). Including this information would improve the figure's scientific rigor and allow for better comparison with other studies.
• Lack of quantitative measurements: The figure does not include quantitative measurements of tumor volume or contrast enhancement. Including these measurements would provide a more objective assessment of the treatment effect and relapse.
• Scale and orientation of images: The scale and orientation of the images are maintained for easy comparison, and the figure is also easy to interpret.

• Visual representation of disease progression: The figure provides a visual representation of the disease progression and relapse in Patient 3. Including both T2/FLAIR and T1 with contrast sequences at diagnosis is useful for characterizing the tumor.
• Caption clarity and key events: The caption is comprehensive, but it would be helpful to highlight the key events at each time point (e.g., initial tumor size, response to treatment, time of relapse).
• Accessibility for non-experts: The figure assumes a certain level of familiarity with MRI interpretation. Annotations highlighting key features such as tumor boundaries, edema, and areas of recurrence would improve accessibility for non-experts.

FIGURE 7 Patient 4: (A) brain MRI on diagnosis (T2/FLAIR) (B) brain MRI on diagnosis (T1 with contrast) (C) 24-month follow-up brain MRI (T1 with contrast) (D) 32-month follow up brain MRI. GBM relapse (T1 with contrast) (E) 41-month follow up brain MRI (T1 with contrast).

• MRI at diagnosis: Figure 7 presents a series of brain MRIs for Patient 4. Panel A shows the brain MRI at diagnosis using T2-weighted Fluid Attenuated Inversion Recovery (FLAIR) sequence, revealing a large heterogeneously enhancing tumor in the left parietal lobe. Panel B shows the brain MRI at diagnosis using T1-weighted sequence with contrast enhancement, highlighting the tumor's vascularity and blood-brain barrier disruption.
• Follow-up MRI characteristics: Panel C shows the 24-month follow-up brain MRI using T1-weighted sequence with contrast, demonstrating the initial response to treatment and a reduction in tumor size. Panel D shows GBM relapse at 32-month follow up using T1-weighted sequence with contrast, revealing the tumor re-growth and increased contrast enhancement. Panel E shows the 41-month follow up brain MRI using T1-weighted sequence with contrast.
• Visual timeline of tumor behavior: The sequential MRIs provide a visual timeline of the patient's condition, demonstrating the initial response to treatment, the subsequent relapse, and the overall progression of the disease. The figure allows for a visual assessment of the tumor's behavior over time.

• Appropriateness of MRI sequences: The figure provides valuable visual evidence of the patient's response and subsequent relapse. The use of T2/FLAIR and T1 with contrast sequences is appropriate for assessing tumor characteristics and response to therapy.
• Missing MRI acquisition parameters: The figure lacks information on the MRI acquisition parameters (e.g., field strength, slice thickness). Including this information would improve the figure's scientific rigor and allow for better comparison with other studies.
• Lack of quantitative measurements: The figure does not include quantitative measurements of tumor volume or contrast enhancement. Including these measurements would provide a more objective assessment of the treatment effect and relapse.
• Scale and orientation of images: The scale and orientation of the images are maintained for easy comparison, and the figure is also easy to interpret.

• Visual record of disease progression and relapse: The figure provides a comprehensive visual record of Patient 4's tumor progression, treatment response, and subsequent relapse. The inclusion of both T2/FLAIR and T1 with contrast sequences at different time points is valuable for demonstrating the changes in tumor characteristics.
• Caption clarity and key events: The caption is detailed, but it could be improved by highlighting the key events at each time point, such as the initial tumor size, the degree of response to treatment, and the timing and characteristics of the relapse.
• Accessibility for non-experts: The figure assumes a certain level of familiarity with MRI interpretation. Annotations highlighting key features such as tumor boundaries, edema, areas of necrosis, and signs of relapse would improve accessibility for non-experts.

FIGURE 7 However, 32 months after diagnosis, the patient suffered a GBM relapse (Figure 7).

• Description of GBM relapse: The caption indicates that 32 months after the initial diagnosis, Patient 4 experienced a recurrence of Glioblastoma Multiforme (GBM). This implies that the patient had initially responded to treatment, but the tumor subsequently regrew.
• Transition in clinical course: The caption serves as a transition point, indicating a change in the patient's clinical course from a period of stability or remission to active disease progression.
• Timeframe of relapse: The 32-month timeframe provides a specific value indicating the duration of the initial response before the relapse occurred. This information is important for assessing the long-term efficacy of the treatment regimen.

• Accuracy of clinical information: The caption accurately reflects the clinical course of the patient, indicating a relapse of GBM after a period of initial response. The 32-month timeframe is consistent with the known patterns of GBM recurrence.
• Missing diagnostic criteria: The caption lacks specific details on the diagnostic criteria used to define relapse. Specifying the imaging and clinical criteria used to determine relapse would improve the caption's scientific rigor.
• Integration with imaging findings: The caption relies on the reader to refer to Figure 7 for visual confirmation of the relapse. Including a brief description of the imaging findings (e.g., new contrast-enhancing lesion) would strengthen the caption's scientific validity.

• Clarity of relapse information: The caption clearly indicates the occurrence of a GBM relapse in Patient 4 at 32 months post-diagnosis. However, it would be more informative to specify the location and characteristics of the relapse, such as size and contrast enhancement.
• Precision of language: The phrase 'suffered a GBM relapse' is somewhat vague. Specifying the criteria used to define relapse (e.g., new contrast-enhancing lesion on MRI) would improve the caption's precision.
• Accessibility for non-experts: The caption assumes that the reader is familiar with the concept of GBM relapse. Briefly explaining what GBM relapse entails (i.e., recurrence of tumor growth after a period of remission) would improve accessibility for non-experts.

FIGURE 8 Patient 5: (A) pre-operative brain MRI (T2/FLAIR) (B) pre-operative brain MRI (T1 with contrast) (C) 34-month follow up brain MRI (T1 with contrast) (D) GBM relapse 40-month follow up brain MRI (T1 with contrast).

• MRI at diagnosis: Figure 8 presents a series of brain MRIs for Patient 5. Panel A shows the pre-operative MRI using T2-weighted Fluid Attenuated Inversion Recovery (FLAIR) sequence, revealing a large heterogeneously enhancing tumor in the right hemisphere. Panel B shows the pre-operative MRI using T1-weighted sequence with contrast enhancement, highlighting the tumor's vascularity and blood-brain barrier disruption. The tumor expands contralaterally through corpus callosum, as well as supratentorially to the right cerebellum.
• Follow-up MRI characteristics: Panel C shows the 34-month follow-up brain MRI using T1-weighted sequence with contrast, demonstrating the initial response to treatment and a reduction in tumor size. Panel D shows GBM relapse at 40-month follow up using T1-weighted sequence with contrast, revealing the tumor re-growth and increased contrast enhancement.
• Visual timeline of tumor behavior: The sequential MRIs provide a visual timeline of the patient's condition, demonstrating the initial response to treatment and the subsequent relapse. The figure allows for a visual assessment of the tumor's behavior over time.

• Appropriateness of MRI sequences: The figure provides valuable visual evidence of the patient's response and subsequent relapse. The use of T2/FLAIR and T1 with contrast sequences is appropriate for assessing tumor characteristics and response to therapy.
• Missing MRI acquisition parameters: The figure lacks information on the MRI acquisition parameters (e.g., field strength, slice thickness). Including this information would improve the figure's scientific rigor and allow for better comparison with other studies.
• Lack of quantitative measurements: The figure does not include quantitative measurements of tumor volume or contrast enhancement. Including these measurements would provide a more objective assessment of the treatment effect and relapse.
• Scale and orientation of images: The scale and orientation of the images are maintained for easy comparison, and the figure is also easy to interpret.

• Visual representation of disease progression: The figure presents a clear visual timeline of Patient 5's tumor, showing the initial presentation and subsequent relapse. Including both T2/FLAIR and T1 with contrast sequences is useful for characterizing the tumor.
• Caption clarity and completeness: The caption is detailed, but it could be improved by including information on the size of the tumor at each time point, the specific location of the relapse, and any other relevant clinical information.
• Accessibility for non-experts: The figure assumes a certain level of familiarity with MRI interpretation. Annotations highlighting key features such as tumor boundaries, edema, areas of necrosis, and signs of relapse would improve accessibility for non-experts.

FIGURE 8 However, in May 2023 (36 months post diagnosis), the patient suffered a GBM recurrence (Figure 8) and was initiated with second line bevacizumab (5-7 mg/kg) and irinotecan (120 mg/m²) every other week.

• Description of GBM recurrence: The caption describes the recurrence of Glioblastoma Multiforme (GBM) in Patient 5, 36 months after the initial diagnosis. This indicates a failure of the initial treatment regimen to provide long-term control of the disease.
• Specification of second-line treatment: The caption specifies the initiation of second-line chemotherapy with bevacizumab and irinotecan. Bevacizumab is a monoclonal antibody that inhibits angiogenesis (blood vessel formation), while irinotecan is a topoisomerase I inhibitor that disrupts DNA replication.
• Details on chemotherapy regimen: The caption provides the dosages of the chemotherapeutic agents (bevacizumab 5-7 mg/kg, irinotecan 120 mg/m²) and the frequency of administration (every other week).

• Accuracy of clinical information: The caption accurately reflects the clinical course of the patient, indicating a recurrence of GBM and the subsequent initiation of second-line treatment. The specified drugs and dosages are consistent with standard clinical practice for recurrent GBM.
• Missing diagnostic criteria: The caption lacks information on the criteria used to diagnose recurrence. Specifying the imaging and clinical criteria used to determine relapse would improve the caption's scientific rigor.
• Integration with imaging findings: The caption relies on the reader to refer to Figure 8 for visual confirmation of the recurrence. Including a brief description of the imaging findings (e.g., new contrast-enhancing lesion) would strengthen the caption's scientific validity.

• Clarity of relapse and treatment information: The caption clearly indicates the occurrence of a GBM recurrence in Patient 5 at 36 months post-diagnosis. It also provides information on the second-line treatment initiated, which is valuable for understanding the patient's management.
• Detail on drug names and dosages: The inclusion of drug names and dosages (bevacizumab 5-7 mg/kg, irinotecan 120 mg/m²) adds a level of detail that is useful for clinicians and researchers. However, briefly explaining the mechanism of action of these drugs would improve the caption's informativeness.
• Accessibility for non-experts: The caption assumes that the reader is familiar with the concept of second-line treatment. Briefly explaining what this entails (i.e., treatment initiated after the failure of initial therapy) would improve accessibility for non-experts.

FIGURE 9 Patient 6: (A) brain MRI on diagnosis (T2/FLAIR) (B) brain MRI on diagnosis (T1 with contrast) (C) 18-month follow up brain MRI (T1 with contrast) (D) 24-month follow-up brain MRI (T1 with contrast).

• MRI at diagnosis: Figure 9 presents a series of brain MRIs for Patient 6. Panel A shows the brain MRI at diagnosis using T2-weighted Fluid Attenuated Inversion Recovery (FLAIR) sequence, revealing a space-occupying lesion on the left occipito-parietal region. Panel B shows the brain MRI at diagnosis using T1-weighted sequence with contrast enhancement, highlighting the tumor's vascularity and blood-brain barrier disruption.
• Follow-up MRI characteristics: Panel C shows the 18-month follow-up brain MRI using T1-weighted sequence with contrast, demonstrating the response to treatment. Panel D shows the 24-month follow-up brain MRI using T1-weighted sequence with contrast, showing the subsequent changes.
• Visual timeline of tumor behavior: The sequential MRIs provide a visual timeline of the patient's condition, demonstrating the response to treatment and any subsequent changes in the tumor's characteristics over time. The figure allows for a visual assessment of the tumor's behavior over time.

• Appropriateness of MRI sequences: The figure provides valuable visual evidence of the patient's tumor. The use of T2/FLAIR and T1 with contrast sequences is appropriate for assessing tumor characteristics.
• Missing MRI acquisition parameters: The figure lacks information on the MRI acquisition parameters (e.g., field strength, slice thickness). Including this information would improve the figure's scientific rigor and allow for better comparison with other studies.
• Lack of quantitative measurements: The figure does not include quantitative measurements of tumor volume or contrast enhancement. Including these measurements would provide a more objective assessment of the treatment effect.
• Scale and orientation of images: The scale and orientation of the images are maintained for easy comparison, and the figure is also easy to interpret.

• Visual record of tumor and follow-up: The figure presents a visual record of Patient 6's tumor, showing the initial presentation and follow-up images. Including both T2/FLAIR and T1 with contrast sequences at diagnosis is useful for characterizing the tumor.
• Caption clarity and key events: The caption is detailed, but it would be helpful to highlight the key events at each time point, such as the size of the tumor at diagnosis and the response to treatment, if any, at follow-up.
• Accessibility for non-experts: The figure assumes a certain level of familiarity with MRI interpretation. Annotations highlighting key features such as tumor boundaries, edema, and areas of enhancement would improve accessibility for non-experts.

FIGURE 9 In February 2022, 22 months after the diagnosis, the patient experienced a focal epileptic seizure and was diagnosed with a recurrence of the GBM on a subsequent brain MRI (Figure 9).

• Description of clinical events: The caption describes the patient's experience of a focal epileptic seizure 22 months after the initial diagnosis, leading to a brain MRI that confirmed GBM recurrence. This indicates a failure of the initial treatment to provide long-term control of the disease.
• Connection between clinical and imaging findings: The caption connects the clinical symptom (focal seizure) to the imaging diagnosis (GBM recurrence), highlighting the importance of correlating clinical findings with radiological evidence.
• Timeframe of recurrence: The caption provides a specific timeframe (22 months) indicating the duration of remission before recurrence. This information is important for assessing the long-term efficacy of the initial treatment regimen.

• Accuracy of clinical information: The caption accurately reflects the clinical course of the patient, indicating a focal seizure leading to the diagnosis of GBM recurrence. This sequence of events is consistent with the known clinical presentation of GBM.
• Missing diagnostic criteria: The caption lacks specific details on the diagnostic criteria used to define recurrence. Specifying the imaging and clinical criteria used to determine relapse would improve the caption's scientific rigor.
• Integration with imaging findings: The caption relies on the reader to refer to Figure 9 for visual confirmation of the recurrence. Including a brief description of the imaging findings (e.g., new contrast-enhancing lesion) would strengthen the caption's scientific validity.

• Clarity of clinical events: The caption clearly indicates that the patient experienced a focal seizure, which led to the diagnosis of GBM recurrence. This provides a clear understanding of the clinical events that led to further imaging.
• Inclusion of timeframe: The caption includes the timeframe of 22 months after the initial diagnosis, which is important for understanding the duration of remission before relapse.
• Accessibility for non-experts: The phrase 'focal epileptic seizure' and 'GBM recurrence' may not be familiar to all readers. Briefly explaining these terms (e.g., focal seizure: seizure affecting one part of the brain; GBM recurrence: return of tumor growth) would improve accessibility for non-experts.

FIGURE 10 Study flow chart. pd, post diagnosis; ECOG, Eastern Cooperative Oncology Group.

• Overview of study flow: Figure 10 presents a study flow chart, which visually summarizes the enrollment and participation of patients in the study. The flow chart begins with 'Eighteen (18) patients with glioblastoma multiforme' and then branches into two groups: 'Patients who followed the diet>6 months' and 'Patients who didn't comply to the KD'.
• Average ages of the groups: The flow chart indicates that out of the 18 patients, the average age of the group who followed the diet >6 months was 54.66 ± 8.6, while the average age of the group who didn't comply to the KD was 55.66 ± 12.9.
• Outcomes for both groups: The flow chart shows that among the patients who followed the diet, there were 1 woman and 5 men. It also shows outcomes such as 'Alive 28 months pd ECOG 0' and '3 alive, 39-41-80- months pd, 2 with ECOG 0, 1 with ECOG 3, 2 passed away 36-43 months pd'. Similar data is shown for the group that didn't comply with the KD.

• Transparency of patient flow: The flow chart provides a clear and transparent representation of the patient flow through the study. This is important for assessing the study's internal validity and identifying potential sources of bias.
• Missing compliance criteria: The flow chart lacks information on the specific criteria used to define 'compliance' with the ketogenic diet. Clearly stating the criteria for compliance (e.g., blood ketone levels, dietary records) would improve the flow chart's scientific rigor.
• Lack of statistical comparisons: The flow chart presents the number of patients in each group, but it does not include statistical comparisons between the groups. Including p-values or confidence intervals for key outcomes would allow readers to assess the statistical significance of the findings.

• Clarity of study design: The flow chart provides a clear and concise overview of the study design and patient enrollment. The use of boxes and arrows effectively illustrates the flow of participants through the study.
• Explanation of abbreviations: The caption provides a brief explanation of the abbreviations used (pd, post diagnosis; ECOG, Eastern Cooperative Oncology Group), which is helpful for readers unfamiliar with these terms.
• Inclusion of reasons for patient drop-out: The flow chart effectively summarizes the key numbers and outcomes of the study. However, it would be useful to include more information on the reasons for patient drop-out (i.e., why patients did not follow the diet), as this is important for interpreting the results.