A man reads a proton therapy brochure. The diagrams are compelling: a beam that enters the body, releases its energy in a precise burst, and stops. No exit dose spilling beyond the tumor. Compared to the photon plan his local hospital offered, the proton plan looks like a scalpel against a chisel. He asks his oncologist: shouldn't he make the drive and pay out of pocket? The physics in the brochure is real. But he has quietly substituted one question for another, and that substitution is what this chapter is about.

Ionizing radiation doesn't know what it's supposed to hit. A beam aimed at a prostate tumor also traverses skin, fat, bowel, bladder. Every radiation side effect comes from healthy tissue absorbing dose it was never meant to absorb. The entire history of radiation oncology is the history of getting better at aiming, concentrating dose where the tumor is and withdrawing it from where it isn't. That ratio is called the therapeutic ratio. Raising it is the goal. Whether any particular technique succeeds at raising it for actual patients is the question that requires trials.

Before a radiation beam reaches the patient, it passes through a multi-leaf collimator, a bank of independently movable metal leaves that can shape the beam into nearly arbitrary patterns. Intensity-modulated radiation therapy uses this device to deliver many beams from many angles, combining into a dose cloud that can wrap around a critical structure. The old rectangular field either underdosed the tumor to spare the spinal cord, or overdosed the cord to treat the tumor. IMRT escapes that tradeoff. The transformative case is head and neck cancer: parotid glands spared, severe dry mouth reduced. That is a clinical benefit, not just a prettier plan.

A perfectly shaped dose distribution is useless if the tumor isn't where the plan assumed it would be. The prostate sits next to a bladder and rectum that vary in filling every day. If the rectum is more distended than on the planning day, the prostate shifts and the intended high-dose region lands in the wrong place. Image-guided radiotherapy addresses this by imaging the patient in treatment position at each fraction before delivery. Cone-beam CT, acquired by a detector on the machine itself, can flag a shift and allow repositioning in the two minutes before beam-on. Implanted gold fiducial markers make the prostate easy to localize. And on an MR-linac, soft tissue is visible in real time, with the beam gated to fire only when the tumor is in position.

SBRT has been extended to oligometastatic disease, patients with one to five metastatic deposits that are amenable to ablative treatment. The SABR-COMET trial randomized patients with controlled primary disease and one to five metastatic lesions to standard palliative care versus stereotactic ablative body radiotherapy of all lesions, and reported a survival benefit in the SBRT arm. This is an evolving area. The biology of oligometastatic disease remains contested: which patients derive genuine benefit from ablation, and which have favorable biology and would do well regardless? Those are live questions without settled answers.

Radiation as a single modality treats the tumor it can see. Systemic therapy treats the disease imaging cannot find. Their combination is not additive, it is synergistic. Concurrent chemoradiation pairs radiation with chemotherapy at the same time: the chemotherapy sensitizes tumor cells, disrupting DNA repair, synchronizing cells into radiation-sensitive phases, and impairing the hypoxic cells that would otherwise be most resistant. The result is greater tumor cell killing per fraction. Cisplatin-based concurrent chemoradiation is standard for locally advanced cervical, head and neck, lung, and esophageal cancers, with survival benefit over radiation alone established in randomized trials. The cost: compounded toxicity, as both modalities damage healthy tissue simultaneously.

The chapter's claim is this: dosimetric superiority is necessary but not sufficient evidence of clinical benefit. The man in the opening case is about to pay a great deal of money for a better dose distribution that has not been shown to buy him a better outcome. The skill being taught is not distrust protons. It's reading a dose distribution and asking: is the tissue I'm saving the tissue that drives my patient's outcome? For a child with a posterior fossa tumor, the answer is yes and the decision goes toward protons. For the man with prostate cancer, the answer is no, and the elegant physics in the brochure and the evidence from clinical trials point in different directions.