This was a dose escalation study in which patients were enrolled to receive 1 of 4 doses in dose ascending order. MTD was defined as the dose level that has an estimated DLT rate nearest to 0.25. This is applicable to the Phase 1 part of the study only. Note that an actual MTD was not reached, however a Phase 2 recommended dose was selected based on the dose that was most tolerable to patients.

All patients who received at least 1 dose of BMX-001, regardless of dose level, after enrolling on study.

This is applicable for Phase 2 only. It was a secondary objective of Phase 1 and that is reported as a separate outcome measure. Assessment of overall survival. With standard treatment, the median survival of Grade IV patients is expected to be 14.6 months, and the median survival of Grade III is approximately 36 months. Given that we anticipate that approximately 10% of patients to be Grade III, we estimate that the overall median survival with standard treatment to be roughly 16.7 months.

All enrolled participants enrolled in Phase 2 were included in the analysis. This was not a primary objective of Phase 1.

Median Overall Survival (OS) is a key clinical outcome measure used to assess the efficacy of BMX-001 in combination with standard radiotherapy (RT) and temozolomide (TMZ) in patients with newly diagnosed high-grade gliomas (HGG). OS is defined as the time from the date of study enrollment to the date of death from any cause. Survival status was assessed at regular follow-up intervals (e.g., every 3 months post-treatment) through medical records, patient contact, and clinical evaluations. The final analysis was conducted after a pre-specified number of events (deaths) have occurred.

Expected Outcome: Prolonged median OS compared to historical or control data (RT + TMZ alone) would suggest a survival benefit associated with BMX-001.

This analysis includes all participants enrolled. A Kaplan-Meier analysis is not appropriate for very small groups in each dose escalation cohort because the statistical power would be insufficient, leading to unreliable survival estimates. According to the study statistical analysis plan, OS will be assessed for all participants enrolled in Phase 1 as a whole, rather than separately by dose level.

This is only applicable for the Phase 1 portion of the study. All patients who received at least 1 dose of BMX-001, regardless of dose level, after enrolling on study are included in the analysis.

All patients who received at least 1 dose of BMX-001, regardless of dose level, after enrolling on study.

The Controlled Oral Word Association Test (COWAT) measures verbal fluency which reflects executive functioning, processing speed, mental flexibility, and language output.

The raw score is then adjusted for age, education, and normative data to allow meaningful comparisons across individuals. The standardized score (T-score) is calculated using normative reference tables.

Higher Scores = Better Cognitive Function, lower scores = cognitive impairment. T score range is 0-100.

A T-score of 50 represents the mean performance in the normative population, this means a score of 50 indicates the population mean performance with a standard deviation of 10.

T-scores below 40 suggest clinically significant cognitive decline. Scores range from < 30 (significantly below average), to 30-39 (below average), 40-59 (average), and greater than or equal to 60 is above average.

Stable or improved COWAT scores over time would suggest that adding BMX-001 may help preserve cognitive function.

Only 6 out of 17 eligible participants completed COWAT assessments at the time this objective was measured 24 weeks after the end of standard of care Radiotherapy + Temozolomide.

Part A of the Trails Making Test (TMT) measures visual attention and processing speed by requiring the patient to connect numbers sequentially (1 → 2 → 3, etc.) as quickly as possible. This is a widely used neuropsychological test that measures visual attention, processing speed, and psychomotor function.

Raw scores are measured in seconds (time to complete the task), but T-scores are derived by converting time-based results into a normed distribution accounting for age and education.

Mean (SD) change in cognitive assessment in T-Scores is reported and the range is 0-100.

T score interpretation and range is < 30 (significantly impaired), 30-39 (mildly impaired), 40-59 (Average), >/= 60 (Above average). A T-score of 50 represents the mean performance in the normative population, this means a score of 50 indicates the population mean performance with a standard deviation of 10.

Higher scores = better performance.

Only 6 out of 17 eligible participants completed Trail Making Test (TMT): Part A assessments at the time this objective was measured 24 weeks after the end of standard of care Radiotherapy + Temozolomide.

The Trails Making Test (TMT) Part B is a widely used neuropsychological test that measures set-shifting ability, processing speed, and working memory. While Part A focuses on speed, visual scanning, and attention.

Prolonged completion time or an increase in errors compared to baseline or age-adjusted norms would indicate cognitive decline.

Raw scores are measured in seconds (time to complete the task), but T-scores are derived by converting time-based results into a normed distribution accounting for age and education.

Mean (SD) change in T score is reported, range is 0-100. T score interpretation and range is < 30 (significantly impaired), 30-39 (mildly impaired), 40-59 (Average), >/= 60 (Above average). A T-score of 50 represents the mean performance in the normative population, this means a score of 50 indicates the population mean performance with a standard deviation of 10.

Higher scores = better performance.

Only 6 out of 17 eligible participants completed Trail Making Test (TMT): Part B assessments at the time this objective was measured 24 weeks after the end of standard of care Radiotherapy + Temozolomide.

Hopkins Verbal Learning Test- Revised (HVLT-R) is a standardized neuropsychological assessment that measures verbal learning and memory, including total (immediate recall). Total Recall T-score is derived from the sum of correctly recalled words across three learning trials and is normalized based on age-adjusted normative data. This provides insight into immediate verbal memory performance and learning ability.

Total Recall (Immediate Memory & Learning) is Sum of correctly recalled words across three learning trials and is a measurement of immediate recall capacity and learning efficiency.

Raw score range is 0-36, T score range is 0-100. T scores >/= 60 are above average and T scores < 40 are below average. A T-score of 50 represents the mean performance in the normative population, this means a score of 50 indicates the population mean with a standard deviation of 10

Only 6 out of 17 eligible participants completed HVLT-R assessments at the time this objective was measured 24 weeks after the end of standard of care Radiotherapy + Temozolomide.

To assess the safety and tolerability of standard RT and TMZ in combination with BMX-001 compared to standard RT and TMZ alone in newly diagnosed HGG patients. The phase 2 portion of this study has two adverse event endpoints:

1. The proportion of patients who experience any grade 3 or 4 adverse event during radiation and temozolomide treatment, and
2. The proportion of patients who experience a grade 3 or 4 adverse event that is definitely, possibly, or probably related to BMX-001 treatment during this same period.

This outcome measure does not apply to the Phase 1 portion of the study, as it was not designated as an outcome for Phase 1 but was specified only for Phase 2.

This analysis focuses on the proportion of patients who experience any grade 3 or 4 adverse event during radiation and temozolomide treatment.

This outcome measure does not apply to the Phase 1 portion of the study, as it was not designated as an outcome for Phase 1 but was specified only for Phase 2.

To assess the safety and tolerability of standard RT and TMZ in combination with BMX-001 compared to standard RT and TMZ alone in newly diagnosed HGG patients.

This outcome is measured to test the proportion of patients who experience a grade 3 or 4 adverse event that is definitely, possibly, or probably related to BMX-001 treatment during this same period.

This outcome measure does not apply to the Phase 1 portion of the study, as it was not designated as an outcome for Phase 1 but was specified only for Phase 2. It also only applies to Arm A in the study as Arm B did not receive the study drug.

This outcome measure does not apply to the Phase 1 portion of the study, as it was not designated as an outcome for Phase 1 but was specified only for Phase 2.

This analysis focuses on the proportion of patients who experience a grade 3 or 4 adverse event that is definitely, possibly, or probably related to BMX-001 treatment during this same period. Data is only presented for Arm A, the BMX-001 treatment group, as subjects in Arm B did not receive BMX-001.

Hopkins Verbal Learning Test- Revised (HVLT-R) is a standardized neuropsychological test that measures verbal learning and memory, including total (immediate recall), delayed recall, and recognition discrimination. It is scored using a T-score which are calculated based on raw scores (e.g., number of words recalled) and standardized using normative data adjusted for age, education, and sometimes sex. T-score range is 0-100. T-scores >/= 60 are above average and T scores < 40 are below average. Higher scores are better. Total recall interpretation - T ≥ 60: above average; T < 40: below average. Delayed recall - T < 30 may suggest memory consolidation issues, and recognition discrimination - T < 30 may indicate impaired recognition memory. A T-score of 50 represents the mean performance in the normative population, this means a score of 50 indicates the population mean performance with a standard deviation of 10.

Mean (SD) change in cognitive assessment is reported.

Only patients that had both a baseline + follow up tests completed could be included in the analysis

This Controlled Oral Word Association Test (COWAT) measure verbal fluency which reflects executive functioning, processing speed, mental flexibility, and language output.

The raw score is adjusted for age, education, and normative data to allow meaningful comparisons across individuals. The standardized score (T-score) is calculated using normative reference tables.

Higher Scores = Better Cognitive Function, lower scores = cognitive impairment. T-score range is 0-100 A T-score of 50 represents the mean performance in the normative population, this means a score of 50 indicates the population mean performance with a standard deviation of 10.

T-scores below 40 suggest clinically significant cognitive decline. Scores range from < 30 (significantly below average), to 30-39 (below average), 40-59 (average), and greater than or equal to 60 is above average.

Stable or improved COWAT scores over time would suggest that adding BMX-001 may help preserve cognition.

Only patients that had both a baseline + follow up tests completed could be included in the analysis

Phase 1 is reported separately. Neurocognitive testing was done and reported here for the Trails Test A and B Part A: Visual attention and processing speed Part B: Executive functioning, task switching, and divided attention. Part A testing time is typically 20-90 sec, and Part B time is typically 40-180 sec.

Raw scores are measured in seconds (time to complete the task), but T-scores are derived by converting time-based results into a normed distribution accounting for age and education.

Mean (SD) change in cognitive assessment in T-Scores is reported T score interpretation and range is < 30 (significantly impaired), 30-39 (mildly impaired), 40-59 (Average), >/60 (above average) A T-score of 50 represents the mean performance in the normative population, this means a score of 50 indicates the population mean performance with a standard deviation of 10.

Lower times = better

Only patients that had both a baseline + follow up tests completed could be included in the analysis

This was not a secondary outcome in Phase 1 and therefore only applies to Phase 2. The proportion of patients who experience grade 3 or 4 thrombocytopenia during concurrent temozolomide and radiation will be recorded within each treatment group. The proportion of patients who experience a platelet count less than 100K during concurrent temozolomide and radiation will also be recorded within each treatment group. For both endpoints described above, a chi-square or Fisher's exact test was conducted to compare the prevalence of such thrombocytopenia observed in patients with and without BMX-001.

For this endpoint to have relevance, a patient must have received some treatment with temozolomide and radiation in either Arm A or B. Analyses suggest that 78 patients in Arm A and 71 patients in Arm B received at least some form of protocol treatment.

The primary analysis of PFS will consider all patients, and consider them in their assigned treatment arm regardless of compliance. This approach to analysis is consistent with an intent-to-treat analysis approach. Progression-Free Survival (PFS) is defined as the time from randomization (Phase 2) or study enrollment (Phase 1) to the first occurrence of either disease progression (as determined by standardized radiographic criteria, the RANO [Response Assessment in Neuro-Oncology] criteria) or death from any cause, whichever occurs first. Patients who have not experienced progression or death at the time of analysis will be censored at their last known follow-up date. PFS will be estimated using the Kaplan-Meier method, providing median PFS and corresponding 95% confidence intervals (CI).

All patients enrolled in each specific arm

The guidelines and criteria for radiographic response will be based on the updated RANO criteria for newly diagnosed GBM. MRI brain with and without contrast will be obtained at enrollment, 2-4 weeks after standard RT and TMZ, and every 8 weeks during adjuvant TMZ. Since this is a study in newly diagnosed patients with HGG, the baseline imaging will be designated as the imaging obtained 2 to 4 weeks after the completion of standard RT and TMZ. At each time point, based on RANO criteria, the subject response will be characterized as Complete Response, Partial Response, Progressive Disease, Stable Disease, or Not Evaluable.