How PET Scans Detect Cancer

01 Cancer Is a Metabolic Signal

A tumor is not merely a lump with a different shape. It is a population of cells that has rewired its behavior: dividing without permission, evading immune surveillance, recruiting blood vessels, and — crucially for PET imaging — consuming energy at an abnormal rate. Most cancers rely heavily on glucose, the body's simplest fuel, because rapidly dividing cells need raw materials and energy faster than normal tissue can provide them. This metabolic appetite is the signal PET scans exploit.

In a clinical PET scan, the tracer is usually fluorodeoxyglucose, abbreviated FDG. FDG is a glucose analog: its molecular structure is close enough to glucose that cells import it through the same membrane transporters and begin processing it through the first steps of glycolysis. But one hydrogen atom has been replaced by fluorine-18, a radioactive isotope. Once FDG is phosphorylated inside the cell, the altered molecule cannot proceed normally through the rest of the metabolic pathway. It becomes trapped, accumulating in tissue in proportion to glucose uptake.

This is why a PET image can reveal cancer before a tumor has produced an obvious anatomical distortion. A CT scan sees density and shape. FDG-PET sees a biochemical behavior. A small cluster of highly active malignant cells can glow as a hotspot even when it is too small or too similar in density to distinguish on conventional imaging. The distinction is not absolute — inflammation and infection also consume glucose — but the metabolic contrast gives oncologists a second, complementary way to find disease.

02 The Radioactive Clock

Fluorine-18 is useful because its half-life is approximately 109.7 minutes: long enough to manufacture FDG, transport it to a hospital, inject it, and image the patient, but short enough that the radioactivity falls rapidly after the scan. Fluorine-18 decays by positron emission. In the nucleus, a proton transforms into a neutron, emitting a positron — the antimatter counterpart of the electron — and a neutrino. The positron travels only a millimeter or two through tissue before encountering an electron.

The collision is an annihilation event. The positron and electron disappear, and their mass becomes energy in the form of two gamma-ray photons, each with an energy of 511 kilo-electron-volts. Conservation of momentum sends the photons in almost exactly opposite directions. A PET scanner surrounds the patient with a ring of detectors made from dense scintillating crystals. When two detectors register 511 keV photons within a narrow coincidence window — typically a few nanoseconds — the system infers that an annihilation occurred somewhere along the line connecting them.

That line is not a pinpoint. The scanner collects millions of these lines of response and reconstructs a three-dimensional distribution of tracer concentration using tomographic algorithms. Corrections account for photon attenuation by the patient's body, random coincidences, detector sensitivity, and physical scatter. The final image is a map of radiotracer uptake, often expressed as a standardized uptake value, or SUV, that normalizes activity for injected dose and body size. SUV is useful, but it is not a direct measurement of malignancy; the number is meaningful only in anatomical and clinical context.

03 Why PET and CT Travel Together

PET is functionally rich but anatomically blurry. A hotspot tells the oncologist that something metabolically unusual is happening, but not always exactly which organ, lymph node, or tissue plane contains it. CT is the opposite: it provides high-resolution anatomy through X-ray attenuation, but most tumors are not uniquely identifiable by density alone. Combining them in a PET/CT scanner solves both problems in one appointment.

The patient lies on a bed that passes first through the CT gantry and then through the PET detector ring. The CT acquisition takes seconds; the PET acquisition takes minutes per bed position. Software registers the two datasets, aligning each functional hotspot with its anatomical location. The result is a fused image in which a bright FDG focus can be localized to a lung nodule, a mediastinal lymph node, a segment of bone, or a distant organ. The CT data also supply the attenuation map needed to correct the PET signal, making the combination more than a side-by-side comparison.

In oncology, PET/CT is used for staging, treatment response, and recurrence. It can reveal whether a cancer has spread to regional lymph nodes or distant organs, information that changes the treatment plan from local surgery to systemic therapy. After treatment, a shrinking mass on CT may contain scar tissue, while persistent metabolic activity may indicate viable tumor. Conversely, a residual PET hotspot can reflect inflammation after radiation or surgery. The technology improves the odds of seeing the whole disease, but it does not replace pathology: a biopsy remains the definitive way to identify cancer cells.

04 The Scan, From Injection to Image

PET imaging begins before the patient enters the scanner. Because insulin and recent food intake alter glucose distribution, patients are typically asked to fast for several hours and avoid strenuous exercise. A technologist measures blood glucose, injects a small amount of FDG through an intravenous line, and allows the tracer to distribute for roughly 45 to 60 minutes. The patient rests quietly during this uptake period; talking, walking, or shivering can activate muscles and create distracting hotspots.

During the scan, the patient lies still on a narrow table while it moves through the detector ring. A whole-body study may take 20 to 40 minutes, depending on the scanner and the number of bed positions. The radioactive dose is small and decays quickly, but the patient emits gamma rays for several hours. Drinking water and urinating help clear unbound tracer through the kidneys. The scan itself is painless except for the injection and, when CT contrast is used, a separate contrast administration.

The raw data are not photographs. They are time-stamped detector coincidences, reconstructed into a volume by iterative algorithms. A radiologist or nuclear-medicine physician reads the images alongside the CT, the patient's history, prior scans, and laboratory results. The interpretation is pattern-based: symmetry, intensity, anatomical plausibility, and change over time all matter. A single bright pixel is not a diagnosis, and a quiet scan does not guarantee the absence of microscopic disease.

05 The False Positives and the Quiet Tumors

FDG is not a cancer detector in the simplistic sense. Activated immune cells also consume glucose, so infection, inflammation, healing wounds, and some benign growths can appear bright. A recent surgery can create a hotspot at the operative site. Granulomatous diseases can mimic metastatic lymph nodes. The brain normally consumes enormous amounts of glucose, which makes FDG-PET naturally bright there and limits its usefulness for many primary brain tumors. The heart, kidneys, and bladder have their own normal patterns of uptake and excretion.

Some cancers are also less visible on FDG-PET. Tumors with low metabolic activity may not accumulate enough tracer to stand out. Certain slow-growing lymphomas, some prostate cancers, renal-cell tumors, and neuroendocrine tumors may require different tracers tailored to their biology. A lesion smaller than the scanner's effective resolution can suffer from the partial-volume effect: its activity is averaged with surrounding tissue, making it appear less intense than it truly is. High blood glucose can also reduce FDG uptake by competing with the tracer.

For these reasons, oncologists do not use a single SUV cutoff as a universal cancer test. They compare uptake with the anatomical finding, the patient's symptoms, the pattern of spread, previous imaging, and — when uncertainty remains — tissue sampling. PET is strongest as a whole-body map of suspicious biology, a way to ask where the disease may be active and whether treatment appears to be working. It is weaker as a standalone answer to what a particular spot is.

06 Beyond FDG: Tracers as Molecular Questions

The most important idea in modern PET is that the scanner is generic but the tracer is specific. Change the radiopharmaceutical and the image asks a different biological question. Fluorine-18 sodium fluoride highlights bone turnover. Gallium-68 DOTATATE binds to somatostatin receptors on many neuroendocrine tumors. Prostate-specific membrane antigen tracers can reveal prostate-cancer deposits that are inconspicuous on FDG. Carbon-11 and other short-lived isotopes enable specialized studies of receptor binding and neurotransmitter systems, though they require a nearby cyclotron or radiopharmacy.

This tracer-centered approach is sometimes called theranostics when diagnosis and treatment are linked. A molecule that binds a tumor receptor can be labeled with a positron emitter for imaging, then labeled with a therapeutic alpha- or beta-emitting isotope to deliver radiation to the same target. The scan becomes not just a picture but a test of whether a patient's tumor expresses the molecular handle required for treatment. Nuclear medicine therefore sits at an unusual intersection: physics supplies the signal, chemistry supplies the targeting, and oncology supplies the question.

07 A Map, Not a Verdict

PET scans detect cancer by translating a biological difference into a physical signal. Malignant cells often consume glucose aggressively; FDG follows that sugar, fluorine-18 decays by emitting positrons, annihilation creates paired gamma rays, and a ring of detectors turns their coincidences into a three-dimensional activity map. Each step is measurable, reconstructable, and vulnerable to its own limits. The final image is not a photograph of cancer. It is a probability-rich map of metabolism.

That distinction is why PET works best in combination. CT supplies the coordinates. MRI can supply exquisite soft-tissue contrast. Pathology supplies cellular identity. Blood tests supply systemic context. PET adds the dimension of what the tissue is doing now. In staging, it can expose a hidden route of spread. In treatment monitoring, it can show metabolic response before anatomy changes. In recurrence, it can distinguish a suspicious pattern from a quiet scar — sometimes. It does not eliminate uncertainty; it relocates uncertainty into a framework clinicians can reason about.