Synopsis
Choose cancer imaging according to the clinical question, sequence diagnosis and staging safely, interpret response and treatment effects, and avoid false precision from incidental or equivocal findings.
- Imaging answers a defined question: detect a lesion, characterise local anatomy, stage spread, guide biopsy, plan treatment, assess complications or measure response.
- Ultrasound, CT, MRI, PET-CT, plain radiography and nuclear medicine have complementary strengths; no modality is a universal gold standard for all cancers.
- Use the tumour-specific first-line staging pathway and image enough anatomy to change management without repeating recent adequate studies.
- A suspicious imaging finding is not always malignant; infection, inflammation, treatment change and benign lesions can mimic cancer or progression.
- Compare response with a documented baseline using consistent technique and recognised disease-specific criteria, not impressionistic size language alone.
- Immunotherapy can cause atypical response and inflammatory appearances; clinical deterioration still requires urgent assessment rather than assuming pseudoprogression.
- Contrast, radiation, pregnancy, renal function, implanted devices and claustrophobia affect modality selection and preparation.
- Urgent imaging for cord compression, obstruction, haemorrhage, thrombosis or treatment toxicity is separate from routine restaging intervals.
Key red flags
New cord, brain, airway, vascular, bowel or urinary compromise requires same-day targeted imaging and acute specialist action.
Investigation priorities
Stage thoracic, abdominal and pelvic anatomy rapidly and assess many acute complications.
Management branches
Cancer is suspected or confirmed and imaging is required.
- State whether the question is diagnosis, local staging, distant staging, biopsy access, emergency complication or treatment response.
- Use the tumour-specific recommended modality and review recent images to avoid duplication, radiation and delay.
- Prepare for contrast, pregnancy, renal, allergy, device and access needs, and escalate if the preferred study is unsafe or unavailable.