A diagnosis names disease.A care system controls decisions.
Oncology converts tissue, anatomy, molecular biology, patient condition and goals into a sequence of time-sensitive decisions. The operating unit is not “cancer” alone. It is a defined malignancy, in a defined person, at a defined extent, with a defined evidence profile and treatment intent.
STATE
Cancer care begins when uncertainty becomes decision-grade evidence.
Oncology spans prevention, detection, diagnostic confirmation, classification, staging, risk stratification, treatment, response assessment, supportive and palliative care, surveillance, survivorship and end-of-life care. Each layer answers a different question; collapsing them creates false certainty.
What is the disease?
Primary site, lineage, histologic type, grade and relevant molecular features.
Where is the disease?
Local invasion, nodal involvement, distant spread and disease burden using the applicable staging system.
Who carries the disease?
Performance status, organ function, comorbidity, frailty, symptoms, prior therapy and preferences.
What is treatment trying to do?
Cure, reduce recurrence risk, control disease, relieve symptoms or preserve function—and how that goal is measured.
A screening abnormality is not a cancer diagnosis. A radiologic lesion is not automatically histology. A detected variant is not automatically a therapeutic target. A stage group is not a complete treatment plan. Every transition must retain provenance: specimen, method, date, report, disease context and accountable interpretation.
Every treatment recommendation should be traceable backward.
Anatomic extent organizes risk. It does not erase biology or personhood.
EXTENT
Before definitive treatment
Built from examination, imaging, endoscopy, biopsy and other pre-treatment evidence.
After resection
Adds surgical and microscopic evidence when applicable; not interchangeable with clinical stage.
After systemic therapy
Residual disease, regression and response descriptors must use the disease-specific framework.
New decision context
Site, timing, prior exposure, resistance and resectability matter more than merely repeating the original label.
Staging systems are cancer-specific and versioned. Hematologic malignancies, central nervous system tumors, pediatric cancers and other diseases may use non-TNM frameworks. A stage value without cancer type, staging system, edition and evidence date is incomplete data.
Multidisciplinary care is not attendance. It is evidence convergence.
| Discipline | Decision-grade input | Question answered | Critical ambiguity | Expected output |
|---|---|---|---|---|
| Pathology | Specimen identity, type, grade, invasion, margins, nodes, biomarker context | What disease is present? | Sampling, heterogeneity, uncertain primary, assay adequacy | Integrated diagnosis with limitations |
| Radiology / nuclear medicine | Distribution, dimensions, invasion, nodal and distant disease, prior comparison | Where is it and how has it changed? | Inflammation, treatment effect, non-measurable disease | Staging and response evidence |
| Surgical oncology | Resectability, morbidity, margin feasibility, reconstruction and nodal strategy | Can meaningful local control be achieved? | Borderline anatomy, functional loss, timing | Procedure or nonoperative rationale |
| Radiation oncology | Target volumes, organs at risk, dose, fractionation, image guidance | Can radiation improve control or palliate safely? | Overlap, motion, prior dose, toxicity constraint | Prescription and planning intent |
| Medical oncology / hematology | Systemic regimen, line, biomarkers, organ function and prior exposure | Which systemic strategy fits this state? | Benefit magnitude, cross-resistance, toxicity | Regimen, monitoring and stop rules |
| Genetics / molecular pathology | Germline indication, somatic result, assay validity and family implications | Is a molecular finding actionable? | VUS, tumor-only ambiguity, clonal hematopoiesis | Qualified interpretation and next test |
| Palliative / supportive care | Symptoms, function, distress, caregiver burden and goals | What suffering or functional loss needs action now? | Late referral, symptom under-reporting | Concurrent symptom and support plan |
Intent is a clinical variable—not a marketing adjective.
Eradicate known disease
May require surgery, radiation, systemic therapy or combinations.
- Define probability and tradeoffs
- Protect dose and schedule integrity
- Plan surveillance and late effects
Modify recurrence risk
Therapy before or after definitive local treatment changes sequencing and evidence interpretation.
- Baseline staging before therapy
- Response or residual disease
- Pathology-treatment reconciliation
Extend life or stability
Benefit must remain proportional to toxicity, burden, alternatives and patient priorities.
- Predefined reassessment
- Resistance and next-line logic
- Stop ineffective therapy
Relieve suffering and protect function
Palliative care can run alongside disease-directed treatment at any stage.
- Symptom target
- Fast, feasible intervention
- Outcome meaningful to patient
“Treatment” is not one thing. Each modality has a different control surface.
A molecular result becomes actionable only after five gates.
Identical gene names can carry different implications by alteration, tumor type, treatment line and regulatory jurisdiction. Variants of uncertain significance do not establish a targeted-treatment indication. Negative testing may reflect true absence, assay scope, specimen quality or low tumor content.
Radiation is prescribed in dose—but controlled in space, time and tissue.
Change on a scan is a signal. Response is an interpreted state.
| Observed state | Possible interpretations | Evidence needed | Unsafe shortcut |
|---|---|---|---|
| Lesion enlarges early | Progression, inflammation, hemorrhage, edema or measurement variation | Therapy type, timing, symptoms, disease-specific criteria and confirmatory evidence | Any enlargement equals treatment failure |
| Marker rises | Progression, flare, inflammation, obstruction or assay variation | Marker validity, trend, imaging and clinical state | Single biomarker overrides the patient |
| ctDNA undetected | Low burden, response, low shedding or assay limitation | Assay scope, tumor biology, timing and orthogonal evidence | Undetected means no cancer |
| New isolated lesion | Metastasis, second primary, benign process or treatment effect | Anatomic pattern, comparison, biopsy feasibility and management consequence | Label without verification when identity changes treatment |
| Symptoms worsen, scan stable | Occult progression, toxicity, comorbidity or functional decline | Focused assessment, labs, alternative imaging and medication review | Stable scan equals stable person |
Adverse-event grading is useful only when it triggers timely action.
Fever during clinically significant immunosuppression, severe breathing difficulty, chest pain, new neurologic deficit, uncontrolled bleeding, confusion, rapidly worsening weakness or other emergency features require urgent local assessment. Patient-specific thresholds and instructions from the treating team take priority.
Twelve pathways. Twelve different decision bottlenecks.
Screen-detected lesion with discordant evidence
- Control point
- Radiology–pathology concordance and receptor-specific evidence.
- Failure
- Treating the imaging category as final biology.
Metastatic nonsquamous carcinoma
- Control point
- Preserve tissue and avoid therapy that compromises a biomarker-directed option.
- Failure
- Starting before critical actionable results when clinically avoidable.
Resected disease with recurrence-risk decision
- Control point
- Separate prognostic from predictive evidence.
- Failure
- Stage label without pathology quality and molecular context.
Locally advanced disease and sequence choice
- Control point
- Anatomic risk, organ preservation criteria and verified surveillance capacity.
- Failure
- Calling clinical complete response without protocolized assessment.
Rising marker after local therapy
- Control point
- Timing, prior treatment field and salvage feasibility.
- Failure
- Marker rise converted directly into one universal treatment.
New lesions during immunotherapy
- Control point
- Distinguish progression, immune-related change and toxicity.
- Failure
- Automatic continuation or discontinuation from one scan alone.
High-grade carcinoma at presentation
- Control point
- Surgical timing, residual disease and hereditary implications.
- Failure
- Molecular result disconnected from family and maintenance decisions.
Potentially curable disease with functional stakes
- Control point
- Oncologic control and speech, swallowing, salivary and airway function.
- Failure
- Planning treatment before supportive baseline work.
FDG-avid disease with uncertain node
- Control point
- Adequate architecture and disease-specific response framework.
- Failure
- Fine-needle sample assumed sufficient for every lymphoma question.
Acute presentation with immediate risk
- Control point
- Time-critical complications and rapid lineage/genetic classification.
- Failure
- Sequential testing that delays stabilization or definitive classification.
Integrated diagnosis after resection
- Control point
- Integrated classification and neurologic function.
- Failure
- Histology alone treated as complete modern classification.
Metastatic disease without obvious origin
- Control point
- Stop low-yield testing when it will not change management.
- Failure
- Unbounded search for origin without decision relevance.
Care continues after treatment—and alongside serious disease.
| Domain | Key evidence | Decision | Failure mode |
|---|---|---|---|
| Recurrence surveillance | Cancer type, stage, treatment, time, symptoms and evidence-based interval | Routine review, targeted test or diagnostic workup | More imaging assumed to mean better surveillance |
| Late effects | Exposures, cumulative dose, field, organ risk and latency | Screen, prevent, manage or refer | Generic follow-up ignores actual treatment exposure |
| Second cancers | Genetic risk, therapy exposure, behavior and standard screening | Risk-adapted prevention and detection | Every symptom attributed to prior cancer |
| Function and return | Fatigue, cognition, mobility, sexuality, fertility, work and participation | Rehabilitation and accommodation | “No evidence of disease” equated with full recovery |
| Palliative care | Symptoms, prognosis understanding, goals, caregiver needs | Concurrent support, advance care planning and escalation | Referral delayed until disease-directed options end |
| End-of-life care | Trajectory, reversibility, preferences, place and support capacity | Proportionate treatment and comfort-focused plan | Emergency decisions without prior goals conversation |
Measure decisions, delivery and outcomes—not only activity.
Oncology and cancer care, defined precisely.
Is a tumor the same as cancer?
No. A tumor is an abnormal mass and may be benign or malignant; some cancers, including many hematologic malignancies, do not form a discrete solid tumor.
Does stage determine treatment by itself?
No. Stage is central, but histology, biomarkers, resectability, prior therapy, organ function, performance, patient goals and disease-specific evidence also shape treatment.
What is the difference between grade and stage?
Grade describes microscopic or biologic features of the tumor; stage describes the extent of disease using the applicable cancer-specific system.
Does an actionable mutation always mean targeted therapy?
No. Actionability depends on the exact alteration, assay validity, tumor type, treatment setting, evidence, regulatory context, patient feasibility and available alternatives.
Is palliative care only for the end of life?
No. Palliative care focuses on symptoms, quality of life, communication and support and can be delivered alongside disease-directed treatment.
Does stable imaging mean treatment is working?
Sometimes, but interpretation depends on treatment intent, cancer type, response criteria, symptoms, timing and other evidence. Stable disease may be meaningful in one context and inadequate in another.
What is a multidisciplinary tumor board?
It is a structured review in which relevant specialists reconcile diagnostic, staging, molecular, treatment and patient evidence into a documented recommendation with uncertainties and ownership.
Is this page medical advice?
No. It is an oncology-system model. Individual diagnosis, urgency, treatment and surveillance require qualified professionals using current disease- and jurisdiction-specific guidance.
Disease-specific evidence before generalization.
Primary starting points include the National Cancer Institute’s cancer biology overview, NCI cancer staging resource, NCI treatment modality index, NCI targeted-therapy listings, the FDA Oncology Center of Excellence and the WHO cancer fact sheet. Treatment selection requires the current cancer-specific guideline, regulatory label, staging edition and local multidisciplinary interpretation.