Scientific Publications

Review of 351 studies on AI/radiomics in immunotherapy: shows median cohort size 101, mostly retrospective, single-centre, RQS median 12/36 — underscores the translational gap in AI for oncology. PubMed+1

In 575 advanced-melanoma patients treated with immunotherapy, a radiomics signature (4 imaging features) achieved AUC ~0.92 for estimating 6-month survival vs ~0.80 for RECIST 1.1. JAMA Network

Radiomics signatures predicted sensitivity in NSCLC to therapies (nivolumab, docetaxel, gefitinib) with AUCs up to ~0.82. PubMed

In 667 metastatic colorectal-cancer patients treated with FOLFIRI ± cetuximab, a radiomics signature measured at baseline and 8 weeks predicted treatment sensitivity and overall survival. PubMed+1

Editorial piece discussing the role of AI in precision oncology, special focus on immunotherapy, imaging biomarkers, and the imperative of clinical translation. PubMed+1

This pivotal Phase 3 randomized clinical trial evaluated a novel 3-antigen HBV vaccine against the standard single-antigen vaccine in 2,838 healthy adults. The study demonstrated the superiority of the 3-antigen vaccine, which achieved significantly higher seroprotection rates after both two doses (90.4% vs. 51.6%) and three doses (99.3% vs. 94.8%). The 3-antigen vaccine also induced a faster and more robust antibody response, establishing its potential to provide quicker and more comprehensive protection against HBV infection.

In this large-scale study of 2,415 adults, a two-dose regimen of an HBV vaccine with a novel immunostimulatory adjuvant (HBV-ISS) was compared to the standard three-dose regimen of Engerix-B. The two-dose HBV-ISS vaccine was found to be statistically superior, achieving a seroprotection rate of 95.1% compared to 81.1% for the standard vaccine. The study concluded that the shorter, two-dose schedule induced a more rapid and superior antibody response, directly addressing the clinical challenge of patient compliance with the traditional six-month vaccination schedule.

This landmark trial pioneered a "patient-initiated" study design where participants self-diagnosed a recurrence of genital herpes and began treatment immediately. The study showed that oral famciclovir was significantly more effective than placebo at reducing the time to lesion healing, cessation of viral shedding, and duration of uncomfortable symptoms. This work established episodic, patient-initiated therapy as a convenient and effective alternative to continuous suppression for managing recurrent genital herpes.

This study investigated whether potent antiviral therapy could restore functional immunity in HIV-infected patients. In 41 patients treated with ritonavir and saquinavir, the therapy not only suppressed viral load and increased CD4+ T-cell counts but also led to significant improvements in functional measures, including T-cell proliferative responses and the production of key cytokines (IL-2, IL-12, IL-10). The research provided crucial evidence that potent viral suppression can lead to true immune reconstitution, demonstrating that HIV-induced immune damage is a potentially reversible process.

This foundational research provided a key insight into the regulation of the anti-inflammatory cytokine Interleukin-10 (IL-10). The study discovered that IL-10 production is differentially regulated in T-cells (enhanced by IL-12 and IL-6) and monocytes (enhanced by TNF-α). This finding described a classic negative feedback loop where pro-inflammatory signals induce their own inhibitor, a fundamental mechanism for maintaining immune system balance that is highly relevant to the rational design of vaccine adjuvants.

This review article addresses a key goal in vaccinology: effective oral vaccine delivery. The authors propose a strategic path forward by actively targeting vaccine antigens to microfold (M) cells, the primary antigen-sampling gateways in the gut. By designing delivery systems that bind to M-cell receptors, the efficiency of antigen uptake could be dramatically enhanced, potentially leading to potent mucosal and systemic immune responses from an oral dose.

This Phase 3 trial evaluated an inactivated vaccine to prevent shingles (herpes zoster) in high-risk cancer patients. The vaccine was found to be well-tolerated and effective in patients with solid tumors undergoing chemotherapy, demonstrating a 63.6% efficacy. However, it was not efficacious in patients with hematological malignancies, a finding that underscores the distinct immunological challenges present in different immunocompromised populations.

This analysis leveraged data from the control arm of the major PATRICIA HPV vaccine trial to precisely map the natural history of HPV infections. The study provided critical estimates for the risk of progression from persistent infection to high-grade cervical lesions, identifying HPV-16 and HPV-33 as carrying a ~25-fold higher risk for progression to CIN3+. This work provides essential data for public health modeling and vaccination strategies.

This review article explores the use of modern computational and bioinformatic tools for the rational design of peptide-based vaccines. It focuses on strategies to overcome the challenges posed by hypervariable viruses (e.g., HIV, influenza) that rapidly mutate and escape traditional immune responses. The paper highlights a forward-looking approach to vaccine development that relies on predictive algorithms and computational biology to identify stable and effective vaccine targets.