PURPOSE Phase I trials are a crucial step in the evaluation of new cancer therapies. Historically, low rates of response (5%) and comparably high rates of death from toxicities (0.5%) have contributed to debates on the ethics and orientation of these trials. With the introduction of novel targeted therapies, a contemporary estimate is needed. METHODS We systematically searched PubMed, Embase, and ClinicalTrials.gov for reports of phase I oncology trials of single-agent targeted immunomodulators, molecularly targeted therapies, and antiangiogenic agents, published between January 2015 and July 2018. Adult and pediatric trials of solid and hematological malignancies were eligible. Treatment-related adverse events (grades 3, 4, and 5) and response rates (objective, complete, and partial) were extracted and analyzed. RESULTS One hundred and fifty-eight trial reports, covering 6,707 patients, were included. The rate of treatment-related deaths was 0.0% (95% CI, 0.0 to 0.1), while 13.2% of patients (9.5 to 17.3) experienced a grade 3 or 4 treatment-related toxicity. The combined objective response rate was 6.4% (4.6 to 8.5). Among trials using tumor biomarkers as eligibility criteria, the objective response rate was higher (12.0% [7.3 to 17.6] compared to 4.9% [2.5 to 5.7], P value < .01). The same was true of trials focusing on a single tumor type (13.4% [8.2 to 19.4]) compared to multiple tumor types (3.8% [2.5 to 5.3], P value < .01). CONCLUSION Reduced grade 5 risk and improved benefit appears to exist in modern phase I oncology trials, particularly in trials that target single tumor types and integrate biomarkers as eligibility criteria. These findings provide information to support informed consent discussions, highlight the need for improved reporting of phase I oncology trials, and provide direction for optimizing their design.

Radiotherapy (RT) is an effective cancer treatment modality, but standard RT often causes collateral damage to nearby healthy tissues. To increase therapeutic ratio, radiosensitization via gold nanoparticles (GNPs) has been shown to be effective. One challenge is that megavoltage beams generated by clinical linear accelerators are poor initiators of the photoelectric effect. Previous computer models predicted that a diamond target beam (DTB) will yield 400% more low-energy photons, increasing the probability of interacting with GNPs to enhance the radiation dose by 7.7-fold in the GNP vicinity. After testing DTB radiation coupled with GNPs in multiple cell types, we demonstrate decreased head-and-neck cancer (HNC) cell viability in vitro and enhanced cell-killing in zebrafish xenografts compared to standard RT. HNC cell lines also displayed increased double-stranded DNA breaks with DTB irradiation in the presence of GNPs. This study presents preclinical responses to GNP-enhanced radiotherapy with the novel DTB, providing the first functional data to support the theoretical evidence for radiosensitization via GNPs in this context, and highlighting the potential of this approach to optimize the efficacy of RT in anatomically difficult-to-treat tumors.

e14060 Background: Phase 1 clinical trials are a crucial step in the evaluation of new cancer therapies. However, critics cite a low rate of response to therapy (5%), together with a not insignificant risk of death from associated toxicities (0.5%), suggesting a risk-benefit ratio that may limit the ethics of inviting patients to participate in Phase 1 trials. With the introduction of novel targeted therapies, a contemporary estimate of the risks and benefits is needed. Methods: A systematic review was conducted. Eligible for inclusion were phase 1 trials in both adult and pediatric populations published between Jan 2015 to July 2018 of targeted immunomodulators, molecularly targeted therapies, and antiangiogenic agents for solid and hematological malignancies. We systematically searched PubMed and Embase. Rates of objective response (complete and partial), stable disease, and Grades 3, 4 and 5 treatment-related adverse events were extracted and pooled. The protocol was prospectively registered in PROSPERO (CRD42018100386). Results: 116 trials (109 adult, 6 pediatric, 1 mixed) met inclusion criteria. The studies reported on nearly 4300 patients (52% male, 48% female), ages ranging from 1 to 90. Most trials reported on molecularly targeted therapies (58%), followed by immunomodulators (33%). The combined overall objective response rate was 6% (95% CI 4% to 8%), with a complete response rate of 0.3% (0.1% to 0.7%) and a partial response rate of 5% (3% to 7%). The rate of stable disease was 34% (30% to 38%). Of the three types of therapies, the objective response rate appeared highest in the molecularly targeted therapies, at 8% (5% to 11%). Overall, the rate of treatment-related deaths was 0.02% (0% to 0.2%). Conclusions: Our results suggest that response rates for single agent, targeted phase I anti-cancer therapies are not materially different from estimates derived in the conventional chemotherapy setting. In an era where providers tout the benefits of precision medicine, this is an important consideration when counselling patients considering participation in phase I trials. Treatment-related death rates were very low, weakening criticism that participation puts patients at great risk. Reporting was highly inconsistent across included studies, highlighting a need to improve the quality of reporting in phase I trials.

No abstract