What could be done
Being tested at scaleClinics actCost to try: small (under $1M)1 year to evidence
Taking 250 mg of abiraterone with a low-fat meal gives the same hormone suppression as the standard 1,000 mg on an empty stomach, so the same drug treats four times as many men for the price.
Szmulewitz et al., JCO 2018: low-fat meal abiraterone
Early evidenceResearch actCost to try: medium ($1M to $50M)4 years to evidence
A randomised trial at Tata Memorial found that adding a very low dose of nivolumab to cheap chemotherapy nearly tripled one-year survival in head and neck cancer patients who could not afford standard immunotherapy; larger trials should confirm which cancers this works in.
Patil et al., JCO 2023: low-dose nivolumab with metronomic chemotherapy
SpeculativeRegulators actCost to try: medium ($1M to $50M)4 years to evidence
The FDA now requires new cancer drugs to prove their dose is the right one rather than the highest tolerated; asking the same question of the twenty best-selling approved drugs could cut doses, side effects and cost at once.
FDA guidance: Optimizing the Dosage of Human Prescription Drugs and Biological Products for the Treatment of Oncologic Diseases (August 2024)
Being tested at scaleClinics actCost to try: small (under $1M)1 year to evidence
Pembrolizumab 400 mg every six weeks is approved and works like 200 mg every three weeks, and it halves the number of clinic visits, infusion chairs and travel days without changing the drug bill.
FDA: pembrolizumab label (Drugs@FDA)
Early evidencePhilanthropy actCost to try: large (over $50M)6 years to evidence
The PERSEPHONE trial showed six months of trastuzumab after surgery is as good as twelve, halving the drug cost; no company will run such trials, so public funders and charities must.
Earl et al., Lancet 2019: PERSEPHONE
Being tested at scalePayers actCost to try: small (under $1M)2 years to evidence
One week of breast radiotherapy in five doses works as well as three weeks in fifteen, but clinics paid per dose lose money by switching; paying one price per course removes the penalty.
Brunt et al., Lancet 2020: FAST-Forward
Early evidenceClinics actCost to try: medium ($1M to $50M)3 years to evidence
Under-the-skin versions of atezolizumab, nivolumab and pembrolizumab take minutes rather than an hour and need no infusion chair, so they can be given by a nurse near home, cutting facility fees and travel.
FDA: Drugs@FDA approved labels
Being tested at scaleClinics actCost to try: small (under $1M)1 year to evidence
Weight-based doses rarely match vial sizes, so the leftover is discarded and still billed; closed-system vial sharing and rounding doses to the nearest vial within 10% eliminate most of that waste.
CMS: Discarded drug refunds (JW and JZ modifiers)