Oncology clinical trials cost estimation

How can we estimate the cost of an Oncology Clinical Trial from the Protocol?

Oncology clinical trials represent a significant investment in the health industry. The costs of such trials can vary widely based on factors such as the number of patients, the type of intervention, and the location of the trial. Gene therapy and cell therapy are much more expensive than traditional interventions such as chemotherapy.

Moore et al wrote in JAMA Internal Medicine in 2020 that the median cost of a clinical trial between 2015 and 2016 was $19 million, but costs can easily range up to 100 times that.[4] For oncology trials, an average per-patient cost was reported to be $59,500 in a report by the Pharmaceutical Research and Manufacturers of America (PhRMA).

As drug development costs continue to rise, the ability to predict and manage them becomes increasingly vital. To that end, the use of machine learning in estimating oncology trial costs has proven to be a valuable tool. One such tool is the Clinical Trial Risk Tool developed by Fast Data Science.

Fast Data Science’s Clinical Trial Risk Tool uses machine learning models to analyse clinical trial protocols, predict the cost of a trial, and identify the possible risks. The prices of each intervention in the protocol, including gene and cell therapy are considered, as well as the number of interventions all play a part in this prediction process. The tool successfully predicts the cost on a per-patient basis.

Example oncology clinical trials cost dataset

It is very difficult to obtain reliable cost data for clinical trials, since this is highly commercially sensitive data. However, some data is in the public domain due to public funding bodies, or freedom of information requests. In addition, a number of academic groups have published aggregate data. For a fee, it is possible to obtain commercial databases of trial costs, but these are very expensive.

We are including below a subset of an oncology cost dataset. Some of this data is available publicly on the Grants Dashboard of the California Institute for Regenerative Medicine.

indication (longer)TechnologyCT.gov URLEnrollmentTrial PhaseTotal CostPer Patient Cost ($PP)
Advanced Myeloid Malignancybiologic drug30Phase 132800010933.3
Blood Cancerbiologic drugNCT034833249Phase 15000000555556
Blood Cancerbiologic drugNCT0392593524Phase 16192579258024
B cell cancers, Leukemiabiologic drugNCT03088878156Phase 1/218292674117261
Blood Cancerbiologic drugNCT0222268826Phase 14179598160754
Colon Cancerbiologic drugNCT02953782112Phase 1/21023404891375.4
Leukemia, Acute Myeloid (AML)biologic drugNCT0324847996Phase 1500000052083.3
Blood Cancer, Solid Tumorsbiologic drugNCT0221640988Phase 1650556873926.9
Breast Cancerbiologic drugNCT00781612720Phase 3-104186
Stage IV Melanomacell therapyNCT0043898411Phase 193616485105.8
Stage IV Breast Cancercell therapyNCT0079103723Phase 1/2223635997233
Non-Small Cell Lung Cancercell therapyNCT008507856Phase 1653850108975
Brain Cancercell therapyNCT02546102414Phase 3539101613021.8
Leukemia, Acute Myeloid (AML)cell therapyNCT03301597146Phase 2431000029520.5
Melanomacell therapyNCT018756534Phase 33000000750000
Blood Cancer, Bone Marrow Transplant and Viral Infectioncell therapyNCT0347521260Phase 1/2482558780426.4
Brain Cancercell therapyNCT0220836292Phase 112753854138629
Brain Cancer, Breast Cancercell therapyNCT0369603039Phase 19015149231158
Multiple Myelomacell therapyNCT03288493180Phase 119813407110074
B cell cancers, Leukemiacell therapyNCT0323385457Phase 111034982193596
Lung Cancercell therapyNCT0354636136Phase 111815315328203
Melanoma, Skin cancercell therapyNCT0324086112Phase 1141442211.17869e+06
Sarcomacell therapyNCT0324086112Phase 14693839391153
HIV-related Lymphoma, HIV/AIDSgene therapyNCT0279747018Phase 1/28414265467459
Prostate cancersmall molecule drug232Phase 2/3296952312799.7
Acute Myeloid Leukemiasmall molecule drug60Phase 1/2116674619445.8
Non-Small Cell Lung Cancersmall molecule drug140Phase 2585228841802.1
Solid Tumorssmall molecule drugNCT0195431648Phase 15683693118410
Histogram of costs of Oncology clinical trials

The Clinical Trial Risk Tool is a valuable resource for the health industry, as it helps to identify and quantify the potential risks associated with a specific trial and maneuver through the budgeting process more confidently. Moreover, it helps with both Phase 1 and Phase 2 oncology trial duration and cost estimation.

One of the big advantages of using machine learning in oncology trial cost prediction is it provides a means to analyse complex and multifaceted data, and learn patterns or trends, much faster than humans. Our oncology clinical trial software can significantly reduce the time it takes to estimate the cost of clinical trials.

Cost your oncology protocol

Estimate cost of oncology trial

Try uploading a completed oncology clinical trial protocol to the Clinical Trial Risk Tool for a cost estimation. You can adjust the cost models for your organisation’s specifications.

The costs associated with clinical trials are immense, but with the aid of advanced technology and machine learning, researchers can get an accurate estimate of a trial’s cost from the protocol. Despite the inherent difficulties in predicting such complex costs, tools like Fast Data Science’s Clinical Trial Risk Tool provide a promising option to navigate this challenging landscape.

Other work on predicting oncology trial costs - and how the researchers got cost datasets

In 2016, Sertkaya et al used aggregate data from Medidata solutions to calculate average trial costs by phase and therapeutic area.[6] Sertkaya et al reported that oncology trials cost an average of $4.5 million in Phase 1, $11.2 million in Phase 2, and $22.1 million in Phase 3. Should you use these figures, we advise to take care to update them according to inflation and consumer price index as they are now out of date.

In 2025, Mulcahy et al used SEC filings to estimate trial costs, and modelled costs on the basis of patient months rather than individual trials.[5] They calculated mean R&D costs of $1.31 billion with standard deviation $1.92 billion for all phases of a drug’s development. The fact that the standard deviation is greater than the mean shows the inherent difficulty we face in modelling trial costs. Mulcahy et al did not report any major change to the average development cost when oncology trials were included from their dataset.

References

  1. Martin, L., Hutchens, M., Hawkins, C. et al. How much do clinical trials cost?. Nat Rev Drug Discov 16, 381–382 (2017). https://doi.org/10.1038/nrd.2017.70
  2. May, Clinical trial costs go under the microscope, Nature, 2019, https://doi.org/10.1038/d41591-019-00008-7
  3. Wood TA and McNair D. Clinical Trial Risk Tool: software application using natural language processing to identify the risk of trial uninformativeness. Gates Open Res 2023, 7:56 doi: 10.12688/gatesopenres.14416.1.
  4. Moore, Thomas J., et al. Estimated costs of pivotal trials for novel therapeutic agents approved by the US Food and Drug Administration, 2015-2016. JAMA internal medicine 178.11 (2018): 1451-1457.
  5. Mulcahy, Andrew, et al. Use of clinical trial characteristics to estimate costs of new drug development. JAMA Network Open 8.1 (2025): e2453275-e2453275.
  6. Sertkaya, Aylin, et al. Key cost drivers of pharmaceutical clinical trials in the United States. Clinical Trials 13.2 (2016): 117-126.

Other clinical trial risk, cost, informativeness, and complexity assessments