Since the Trump Administration took office in the White House, there has been an uptick in the number of guidance documents released to assist sponsors in running clinical trials.
While many are broad, there have been some oncology-specific US Food and Drug Administration (FDA) guidance documents, including on improving enrolment, while guidance for rare diseases has significant crossover for oncology, as around one in five (about 20% to 25%) of all cancer diagnoses are classified as rare diseases.
Go deeper with GlobalData
Discover B2B Marketing That Performs
Combine business intelligence and editorial excellence to reach engaged professionals across 36 leading media platforms.
Speaking to Clinical Trials Arena at the European Society of Medical Oncology (ESMO) Congress 2026, which took place from 23 to 27 October in Madrid, Spain, Dr Harpreet Singh, CMO of Precision For Medicine and former division director of oncology at the FDA, said that these shifts signify a period of change that has been needed and developing for several years, but questions remain about how they will work in practice if sponsors are to use these pathways.
This interview has been edited for length and clarity.
Abigail Beaney (AB): We have seen a lot of advances in the US for clinical trials in the past couple of years. What would you say have been the most significant advances for the oncology sector over the past 18 to 24 months?
Harpreet Singh (HS): Operation Trialblazer is exciting. It reflects the FDA’s effort to accelerate early drug development in the US, drawing on models in countries such as China and Australia. Trialblazer most closely resembles Australia’s third-party notification system.
At a broader level, the initiative is about reframing risk. Patient safety is a primary FDA goal, but decades of accumulated regulation can make some preclinical requirements before a first-in-human trial cumbersome and potentially overstate risk. That can delay US patients’ access to novel therapies and encourage sponsors to begin early trials in places such as Australia or China.
Trialblazer aims to place more responsibility with qualified research institutions. Precision for Medicine is applying to become one. I expect small and mid-sized biotech companies developing complex therapies could benefit most.
This is version one of Trialblazer, and I expect it to evolve as the pilot progresses.
AB: How can oncology sponsors balance the FDA real-time clinical trials guidance with patient safety, given the potential for serious adverse events?
HS: I had the same question when I saw the announcement. The programme has not yet been operationalised, and regulators and sponsors will need to strike an important balance.
In late-stage oncology trials, patients may be willing to accept a certain level of risk, particularly when they have refractory, incurable cancers. Early safety data may not show the long-term efficacy – for example, the benefits of immunotherapy – that could affect how patients and clinicians weigh that risk.
One drug that comes to mind is Yervoy (ipilimumab), which was the first CTLA-4 inhibitor to be approved. Early administrations were associated with potentially life-threatening pneumonitis and colitis. Yet what we have seen now is that, when used in combination with checkpoint inhibitors, there have been survival benefits, including benefit to the tail of the curve. If the drug had been deemed a failure based only on those early toxicity signals, that benefit might have been missed.
AB: So, should sponsors take early toxicity signals seriously and manage them – for example, through protocol amendments – while considering whether it is appropriate to continue the study and assess potential longer-term benefits?
HS: Yes. We should also ask what problem real-time data is intended to solve. One perceived bottleneck is the time it takes after a trial reads out to aggregate and clean the data before sending it to the FDA. Real-time data collection during a trial may not solve that problem, because data still has to be aggregated, cleaned, and queried before it is regulatory grade. I’m interested to see how the pilot will work in practice, including the mechanics of sharing data.
AB: Which aspects of the recent FDA rare disease guidance have caught your attention for oncology trials?
HS: I’m very interested in this area. Rare diseases were within my purview when I led an oncology division, and the FDA’s work to accelerate rare disease development has been underway for a long time. The Center for Drug Evaluation and Research has had its Accelerating Rare Disease Cures initiative and a five-year strategic plan.
There are three areas to me that stand out. Firstly, trial design. Can a single-arm trial be used, or is a randomised trial needed? The answer differs by disease. It will be important to understand how the FDA makes these decisions and harmonises its approach. Secondly, endpoints. For diseases that have been studied very little, the FDA is working to identify endpoints that matter to patients and caregivers and can meet regulatory standards. For a composite endpoint, I think there should be a combination of objective measures and subjective, patient- or caregiver-reported measures.
Finally, natural history studies and external controls, which I am very excited about. When a randomised trial is infeasible, it is important to understand the disease’s natural history and evaluate whether those data can be used alongside the evidence generated in clinical studies.
AB: Will the recent FDA guidance, Operation Trialblazer, and other HHS initiatives be enough to bring early-stage oncology research back to the US?
HS: It’s good that HHS recognises the challenges are ecosystem-wide and do not sit solely with the FDA. But I don’t think these initiatives alone will address the long-term impact.
China began a 10-year plan in 2015 to become a major force in biotechnology and drug development. It invested over the long term in clinical-trial infrastructure and clinics, created a 1,000-talent programme that placed talent in the National Medical Products Administration (NMPA), and put government funding and organisational discipline into public-private partnerships.
The US can learn from Australia and China, but the situations differ. China was building from the ground up, while the US already has many ways of conducting early-phase trials. The US will need to improve different parts of its system, build on what works, and change what does not. Government investment in talent and innovative research is also needed, including dedicated funding for the National Cancer Institute. Innovation should cross borders.
AB: What global trends in oncology do you think will have the greatest impact over the next five to ten years?
HS: I believe we will see the era of personalised medicine. Beyond the Moderna–Merck data, biomarker-driven drug development will be important; it is central to Precision for Medicine’s work.
A recent example is dostarlimab for locally advanced rectal cancer in patients whose tumours are mismatch repair deficient or microsatellite instability-high. This group represents about 10% to 15% of rectal cancer patients. Standard treatment for locally advanced rectal cancer often involves surgery, chemotherapy, and radiation, which can have substantial effects on bowel and urinary function. For patients with the relevant biomarker, immunotherapy alone may be an option.
Personalised cancer vaccines are another example. In the Moderna–Merck data, each patient had a different sequence because the vaccines were created based on individual tumours.
Patient selection is also increasingly important in lung cancer, where the disease is being divided into smaller molecular subgroups – some representing 2% or 3% of patients – such as those associated with MET exon 14 skipping, RAS, or HER2. Regulators are increasingly looking for benefit-risk profiles to be strengthened through appropriate patient selection. The goal is the right patient, the right drug, and the right time – and biomarkers help connect those elements.
