promising

Cachexia-directed therapy treats cancer wasting by blocking GDF-15, a hormone that rises in advanced cancer and acts on the brainstem to suppress appetite. Pfizer's antibody ponsegromab improved weight in a phase 2 trial and is in phase 2/3 in pancreatic cancer cachexia; whether weight gain translates into function is the open question.

Giving the same drug at a different time of day, because the body clock changes how much damage it does and how well the immune system responds.

Instead of testing blood every three months, sampling constantly, so a relapse is caught the week it starts.

Designing a protein from scratch on a computer to grip a chosen target, instead of finding one in an animal or a library.

Using a model of what would have happened to a patient on standard treatment, so fewer people have to be randomised to it.

Loading the tiny vesicles cells naturally use to talk to each other with a cancer drug, so the body treats the carrier as its own.

Destroying a tumour mechanically with sound, rather than heat, leaves the debris intact enough for the immune system to learn from it.

Treating one tumour so aggressively that the immune system learns to attack every other one, using the tumour itself as the vaccine.

Lattice radiotherapy deliberately treats a bulky tumour unevenly, placing peaks of tumour-destroying dose at spaced points inside it while the tissue between receives far less, relying on bystander and immune effects to extend the kill. It runs on standard linear accelerators, but evidence is mostly palliative and single-arm, and the mechanism is unsettled.

Cells or drugs that fire only when two conditions are true at once, so healthy tissue expressing just one of them is spared.

Changing the gut bacteria of a patient whose immunotherapy stopped working, in the hope of restarting the response.

Molecular glues are small molecules that stick two proteins together so the cell destroys one of them. They are smaller and more drug-like than bifunctional degraders.

Building a trial around one patient, or letting one trial swap drugs in and out as evidence accumulates.

Organoid-guided therapy means routinely growing a piece of each patient's tumour and testing drugs on it before choosing, rather than relying on genetics alone.

Nanoparticles that turn ordinary radiotherapy X-rays into a much bigger dose exactly where they sit.

Adding a PARP or ATR inhibitor to a radioactive drug so the tumour cannot repair the damage the radiation causes.

RNA drugs that copy themselves inside the cell, or are made as a loop so they last longer. Both aim to get more protein from a smaller dose.

A drug that does nothing until ultrasound hits it, then kills the cells that took it up. Being tested in brain tumours because sound reaches where light cannot.

Choosing treatment from a map of where each cell type sits in the tumour, not just from a list of its mutations.