Oncology Pathway

The foundational layer of any pathway is the curated medical knowledge derived from randomized controlled trials, meta-analyses, and national guidelines (e.g., NCCN, ASCO). This content defines the standard of care for each cancer subtype and stage.

• Regimen Libraries: Structured data defining drug names, doses, frequencies, and durations.
• Decision Logic: Encoded rules that map patient-specific attributes (e.g., biomarker status, ECOG performance score) to the appropriate treatment arm.
• Reference Linking: Direct citations to the published literature supporting each recommendation, enabling clinicians to verify the provenance of the guidance.

Pathways are not one-size-fits-all. A core component is the branching logic that personalizes the care journey based on discrete patient variables. This transforms a static guideline into a dynamic, patient-specific care map.

• Staging Data: Integration of TNM classification (Tumor, Node, Metastasis) to determine disease extent.
• Biomarker Status: Branching based on molecular markers like PD-L1 expression, EGFR mutations, or HER2 amplification.
• Clinical Factors: Incorporation of comorbidities, organ function (e.g., creatinine clearance), and performance status to assess fitness for aggressive therapy.

The pathway explicitly defines the temporal relationship and sequencing between different treatment modalities. This component manages the complex handoffs between surgery, radiation, and systemic therapy.

• Neoadjuvant vs. Adjuvant: Clear logic dictating whether systemic therapy should be administered before or after surgical resection.
• Concurrent Therapy: Rules governing the safe administration of concurrent chemoradiation, including overlapping toxicity management.
• Cycle Tracking: A mechanism to track the number of completed chemotherapy cycles and trigger progression to the next phase of care, such as maintenance therapy or surveillance.

For a pathway to be effective, it must be embedded directly into the clinical workflow via a CDSS. This component represents the technical integration that delivers the right information at the right time.

• EHR Embedding: Launching the pathway from within the oncologist's workflow, often via a SMART on FHIR application.
• Order Set Generation: Automatically translating a selected pathway branch into a signed, protocol-driven chemotherapy order set.
• Rule-Based Alerts: Firing real-time alerts for contraindication checks, dosage range checking, and required pre-medications before treatment is administered.

A critical component is the system's ability to capture when and why a clinician chooses to deviate from the recommended pathway. This closed-loop analytics drives continuous improvement.

• On-Pathway vs. Off-Pathway: Classifying every treatment decision as adherent or non-adherent to the standard pathway.
• Reason Capture: Structured data fields for clinicians to document the rationale for deviation, such as patient refusal, clinical trial enrollment, or a unique comorbidity.
• Outcomes Correlation: Linking pathway adherence data to real-world outcomes like progression-free survival and adverse event rates to refine future pathway versions.

Modern oncology pathways incorporate a financial stewardship layer to support value-based care models. This component balances clinical efficacy with resource utilization.

• Formulary Check: Automated verification that selected drugs are on the payer's approved formulary.
• Therapeutic Substitution: Suggesting biosimilar or therapeutically equivalent agents when clinically appropriate to reduce cost without compromising outcomes.
• Total Cost of Care: Aggregating the projected cost of the entire pathway, including drugs, supportive care, and anticipated hospitalizations, to inform shared decision-making.

A computer-based system that analyzes patient-specific data and provides evidence-based assessments to clinicians at the point of care. In oncology, CDSS platforms integrate pathway rules to recommend optimal chemotherapy sequencing based on cancer stage, biomarkers, and comorbidities.

• Combines patient data with clinical knowledge bases
• Generates real-time, actionable recommendations
• Reduces unwarranted clinical variation across care teams

An HL7 standard for encoding medical knowledge as Medical Logic Modules (MLMs) — independent, situation-action rules. Oncology pathways can be formalized as MLMs that trigger when specific cancer staging data or lab values are entered, ensuring consistent protocol application.

• Encodes if-then clinical logic in a shareable format
• Supports institution-specific pathway customization
• Maintains an audit trail of rule execution for compliance

A Fast Healthcare Interoperability Resources module that standardizes the representation of clinical knowledge artifacts including rules, order sets, and quality measures. This module enables oncology pathways to be expressed as computable, interoperable artifacts that can execute across different EHR systems.

• Defines Clinical Quality Language (CQL) for pathway logic
• Supports reusable knowledge artifact libraries
• Enables pathway sharing across health systems

A decision-making tool that combines multiple clinical predictors — such as tumor size, lymph node status, and genetic markers — to estimate the probability of outcomes like recurrence risk. These rules underpin many oncology pathway branching decisions.

• Integrates prognostic and predictive factors
• Stratifies patients into risk-based treatment groups
• Validated against large cohort studies for reliability

A glass-box interpretable model that combines gradient boosting performance with the intelligibility of generalized additive models. In oncology pathways, EBMs provide transparent risk predictions where clinicians can inspect exactly how each feature — such as tumor grade or Ki-67 index — contributes to a treatment recommendation.

• Offers full model transparency for high-stakes decisions
• Prevents black-box opacity in pathway logic
• Satisfies regulatory requirements for algorithmic explainability

A set of statistical methods for analyzing time-to-event data, such as overall survival or progression-free survival. Oncology pathways rely on survival analysis outputs from clinical trials to define the expected benefit of each treatment sequence.

• Uses Kaplan-Meier estimators and Cox proportional hazards models
• Accounts for censored data in clinical trial populations
• Informs pathway tiering by expected survival advantage