Central nervous system tumours represent one of the most lethal cancer types, particularly among children¹. Primary treatment includes neurosurgical resection of the tumour, in which a delicate balance must be struck between maximizing the extent of resection and minimizing risk of neurological damage and comorbidity^2,3. However, surgeons have limited knowledge of the precise tumour type prior to surgery. Current standard practice relies on preoperative imaging and intraoperative histological analysis, but these are not always conclusive and occasionally wrong. Using rapid nanopore sequencing, a sparse methylation profile can be obtained during surgery⁴. Here we developed Sturgeon, a patient-agnostic transfer-learned neural network, to enable molecular subclassification of central nervous system tumours based on such sparse profiles. Sturgeon delivered an accurate diagnosis within 40 minutes after starting sequencing in 45 out of 50 retrospectively sequenced samples (abstaining from diagnosis of the other 5 samples). Furthermore, we demonstrated its applicability in real time during 25 surgeries, achieving a diagnostic turnaround time of less than 90 min. Of these, 18 (72%) diagnoses were correct and 7 did not reach the required confidence threshold. We conclude that machine-learned diagnosis based on low-cost intraoperative sequencing can assist neurosurgical decision-making, potentially preventing neurological comorbidity and avoiding additional surgeries.

BRAF genomic alterations are the most common oncogenic drivers in pediatric low-grade glioma (pLGG). Arm 1 (n = 77) of the ongoing phase 2 FIREFLY-1 (PNOC026) trial investigated the efficacy of the oral, selective, central nervous system–penetrant, type II RAF inhibitor tovorafenib (420 mg m^{− 2} once weekly; 600 mg maximum) in patients with BRAF-altered, relapsed/refractory pLGG. Arm 2 (n = 60) is an extension cohort, which provided treatment access for patients with RAF-altered pLGG after arm 1 closure. Based on independent review, according to Response Assessment in Neuro-Oncology High-Grade Glioma (RANO-HGG) criteria, the overall response rate (ORR) of 67% met the arm 1 prespecified primary endpoint; median duration of response (DOR) was 16.6 months; and median time to response (TTR) was 3.0 months (secondary endpoints). Other select arm 1 secondary endpoints included ORR, DOR and TTR as assessed by Response Assessment in Pediatric Neuro-Oncology Low-Grade Glioma (RAPNO) criteria and safety (assessed in all treated patients and the primary endpoint for arm 2, n = 137). The ORR according to RAPNO criteria (including minor responses) was 51%; median DOR was 13.8 months; and median TTR was 5.3 months. The most common treatment-related adverse events (TRAEs) were hair color changes (76%), elevated creatine phosphokinase (56%) and anemia (49%). Grade ≥3 TRAEs occurred in 42% of patients. Nine (7%) patients had TRAEs leading to discontinuation of tovorafenib. These data indicate that tovorafenib could be an effective therapy for BRAF-altered, relapsed/refractory pLGG. ClinicalTrials.gov registration: NCT04775485.

Background: Pregnancy affects the pharmacokinetics of most protease inhibitors. Saquinavir, when administered in a tablet formulation, has not been studied extensively in this setting. Methods: A pharmacokinetic, prospective, multicentre trial of HIV type-1-infected pregnant women treated with saquinavir (500 mg tablets) boosted with ritonavir at a dose of 1,000/100 mg twice daily plus a nucleoside backbone was conducted. Pharmacokinetic curves were recorded for 12 h in the second trimester (week 20 ±2), the third trimester (week 33 ±2) and post-partum (weeks 4–6). Blood was sampled pre-dosing and at 1, 2, 3, 4, 6, 8, 10 and 12 h post-dosing. Pharmacokinetic parameters were calculated using WinNonlin software version 4.1. Results: A total of 37 women were included in the analysis. Mean (±sd) values for saquinavir area under the curve (AUC _0–12h ) were 23.47 h•mg/l (11.92) at week 20 ( n=16), 23.65 h•mg/l (9.07) at week 33 ( n=31) and 25.00 h•mg/l (11.81) post-partum ( n=9). There was no significant difference in the saquinavir AUC _0–12h when comparing the data during pregnancy and post-partum. Subtherapeutic plasma concentrations of saquinavir (defined as <0.10 mg/l) were not observed throughout the study. No major safety concerns were noted. Conclusions: Saquinavir exposure in the new tablet formulation generates adequate saquinavir concentrations throughout the course of pregnancy and is safe to use; therefore, no dose adjustment during pregnancy is needed.

To study the pharmacokinetics and short-term efficacy of low and standard dose lopinavir/ritonavir and saquinavir combinations in Thai, human immunodeficiency virus (HIV)-infected, treatment-naive patients. In this open-label, 24-week, prospective study, 48 treatment-naive patients were randomized to lopinavir/ritonavir 400/100 mg+saquinavir 1000 mg twice daily (arm A), lopinavir/ritonavir 400/100 mg+saquinavir 600 mg twice daily (arm B), lopinavir/ritonavir 266/66 mg+saquinavir 1000 mg twice daily (arm C), or lopinavir/ritonavir 266/66 mg+saquinavir 600 mg twice daily (arm D). A 12 h. pharmacokinetic profile in all patients was performed. Plasma concentrations of saquinavir and lopinavir were determined using an HPLC technique. HIV-1 RNA was measured over 24 weeks. Forty-three subjects were included in the pharmacokinetic analysis. The total exposure differed significantly for the different arms. Median values for lopinavir area under the curve at 0–12 h were 128.2, 119.2, 66.1 and 68.5 mg·h/L for arms A–D, respectively. For saquinavir, the median values were 36.9, 19.2, 25.3 and 12.4 mg·h/L for arms A–D, respectively. The proportion of patients having a viral load below 50 copies/mL at week 24 was 39% for arm A, 63% for arm B, 55.0% for arm C, and 69% for arm D. The pharmacokinetic parameters for the different treatment arms were adequate. However, the proportion of subjects with an undectable viral load at week 24 was lower than anticipated.

Background: Generic drugs can contribute to access to treatment for HIV-infected patients. However quality and safety remains an issue of concern. Therefore, we evaluated minimal plasma concentrations and short-term safety of a generic lopinavir/ritonavir 200/50 mg tablet formulation. Methods: In a single-centre prospective pilot study, patients receiving protease-inhibitor-based antiretroviral treatment were switched to a generic lopinavir/ritonavir tablet at the standard dose (400/100 mg twice daily). Minimum drug concentrations (C _min ) of lopinavir and ritonavir were performed before switching (in 16 patients who were on Kaletra® soft-gel capsules) and after 4 weeks (in all patients). Plasma levels of lopinavir and ritonavir were determined by a validated HPLC method. Either the Wilcoxon signed-rank or Mann–Whitney U test was used to compare the groups. Results: A total of 37 patients (18 females) were included in the study. Two stopped their study medications prematurely because of intolerance. The median (interquartile range) lopinavir C _min was 7.2 mg/l (5.8–8.3) and no patients had subtherapeutic levels <1.0 mg/l. No significant difference of lopinavir C _min levels was found between Kaletra®, and the generic product ( P=0.224). By contrast, the C _min of generic ritonavir was higher ( P=0.012). Food did not affect the drug levels. Mild gastrointestinal complaints were reported in 12 patients. Conclusions: The generic lopinavir/ritonavir tablet showed C _min plasma concentrations similar to what is described for the branded product, with good stability, independent of food intake. These data support the efforts in scaling up access to generic second-line treatment in middle- and low-income countries.

Currently, the optimal boosting dose for saquinavir is unknown. Therefore, we evaluated the pharmacokinetics profiles in a cross over setting comparing saquinavir/ritonavir 1500/50 mg (plus NRTI backbone) to saquinavir/ritonavir 1500/100 mg in the same HIV-infected, Thai individuals. The 50% reduction of ritonavir boosting did not result in a change in the pharmacokinetics of saquinavir, whereas the ritonavir exposure was significantly lower when a dose of 50 mg was administered.