A graduate student presents a poster. Her nanoparticle carries doxorubicin, targets a breast cancer receptor, and in cell culture kills tumor cells at one-tenth the usual dose. Then a reviewer asks one question: what percentage of your injected particles actually reached the tumor? She doesn't know. Nobody measured it. Cancer nanomedicine is not the study of clever particles. It is the study of where a particle goes and what it does when it gets there — and that requires measurement, not assumption.

Conventional chemotherapy is a blunt tool. The bloodstream distributes drug throughout the body — bone marrow, gut lining, hair follicles, immune cells — and toxicity from this collateral exposure is what limits the dose a patient can receive. The fundamental promise of cancer nanomedicine is to concentrate drug at the tumor and spare healthy tissue. That promise is real. But delivering on it requires solving a chain of physical problems, and most particles in the literature fail that chain quietly.

The ten-to-two-hundred nanometer range enables behaviors that free drug molecules cannot have. Tumor blood vessels are leakier than normal vasculature — particles in this size range escape through gaps and persist there because tumors drain poorly. This is the enhanced permeability and retention effect, or EPR. Coat a particle with polyethylene glycol and it evades the macrophages that would otherwise clear it, extending its time in circulation from minutes to hours. Encapsulate a therapeutic RNA and it survives the blood long enough to reach a cell. Pack a targeting ligand and an imaging tracer into the same object and you can track delivery and treat simultaneously. These are the engineering arguments for the field.

Before a particle can be claimed to do anything, its physical properties must be measured. A nanoparticle batch is not a single identical molecule — it is a population with a distribution of sizes, surface chemistries, and drug loadings. Polydispersity is not a cosmetic quality metric; it predicts in-vivo behavior, because a fifty-nanometer particle and a three-hundred-nanometer particle are cleared by different organs and extravasate with different efficiencies. Surface chemistry matters because a PEG coating that has aggregated, or a targeting peptide that detached in storage, is invisible to a cell-killing assay but decisive in an animal. Characterization is the step that connects 'we built this object' to 'we know what we built.'

The physics of the enhanced permeability and retention effect are real: leaky tumor vessels plus poor lymphatic drainage means particles in the right size range can accumulate passively. But the field has learned to treat EPR with caution. The widely cited 0.7 percent delivery figure comes from preclinical mouse data, and human EPR may differ substantially. Mouse tumor models develop different vascular architecture than human tumors, often with more pronounced leakiness. Whether EPR is even consistent enough across human patients to be a reliable engineering design target is one of the field's central contested questions — and the chapter addresses it honestly rather than assuming the mouse-to-human translation holds.

Here is the chapter's central claim. Nanomedicine is a delivery discipline. The binding constraint is not particle elegance — it is confirmed biodistribution. The field's recurring error is inferring delivery from downstream response: cell death in culture, tumor shrinkage in mice. Neither closes the dose-loss chain. Only tracking the particle itself closes the chain. What would force revision: a large, well-controlled clinical study where particle design alone, independent of confirmed biodistribution, reliably predicted outcome. The existing pattern is the opposite — the variability of tumor delivery is precisely what makes outcomes unpredictable. And still open: whether a companion diagnostic could select patients by tumor permeability before treatment — changing the field from building better particles to selecting better patients.