Evidence Gap Identification & RWE Study Design: Alexion Preworkshop Material

The GCC Evidence-to-Market (E2M) Framework
A 11-Step Protocol for Rare Disease RWE & Consensus.


Step 1: What You Think Is an Evidence Gap

A research question is not always an evidence gap. An evidence gap is a void in knowledge that, if filled, would inform a key decision for a stakeholder (e.g., a regulator, payer, or clinician). We will explore this distinction during the workshop.

A) Give a General Idea: The First Thing That Comes to Mind

Let’s start by capturing your initial thoughts. Don’t worry about perfection; this is about getting the raw idea down.

Component Your Entry
Disease Area
Example: PNH, gMG, HPP, etc.
Product Name (if applicable)
Example: Soliris, Ultomiris, Strensiq
Lifecycle Stage
Your General Idea or Hypothesis

“I’ve noticed that patients in our region initiate therapy later than in EU registries.”

“We’re often asked whether switching from IV to SC impacts QoL in real-life settings.”

“There’s no local data on pediatric gMG outcomes — even globally, it’s scarce.”
Source of the Idea
(Optional)
Key Message You’re Hoping to Support or Explore
(Optional)

“This idea could help us strengthen the case for earlier treatment”

“We suspect that treatment persistence is a gap in our MENA region”

B) How You Can Categorize It

Tick one or more categories that your idea falls into. We will provide more examples and explanations during the workshop. Items that naturally touch more than one bucket are noted as cross-cutting.

1) Public Health/ Epidemiological

2) Treatment Patterns

3) Clinical Outcomes

4) Safety

5) Economic (Value, Access, System Impact)


Step 2: Literature Search Process

You are likely familiar with PubMed, and your internal teams can support the search process. However, be aware of publication bias; you might not always find the whole story in published literature. That’s why we will also explore Google Scholar, Embase, and ClinicalTrials.gov.

use the table below to document your findings. The goal is to extract the key findings for each database and then list 2 potential evidence gaps based on what you found, or what you didn’t find.

2A. Search Table: One Row per Source

Database Used Search Method Used (Boolean, MeSH, EMTREE, etc.) Exact Search String Used Date of Search Total Results Found Relevance of Results (tick one) Key Notes or Observations
PubMed





Embase (via EMTREE)





Google Scholar





ClinicalTrials.gov





Cochrane Library





Other (specify):





Potential Evidence Gaps Identified

Based on your literature search, list two potential evidence gaps.




Step 3: Evidence Gap Mapping

Use this table (also known as a Gap Matrix) to organize the studies you found during your literature review. Each row can represent a study you found. The goal is to visualize the landscape of existing evidence. The empty cells across countries, endpoints, or populations often represent evidence gaps.

Evidence Mapping Table (Gap Matrix)

Study Type (tick or fill) Study Subcategory Country / Region Patient Population Endpoint Studied Citation(s) or Source
Example subcategories:

Single-center, multi-site, cross-sectional, interviewer-led, claims-based, chart abstraction, KAP survey, exploratory analysis, payer-focused, etc.

Step 4: Formulate PICO with Clinical Precision

Transform the general idea and identified gap into a clinically precise research question using the PICO format. This creates alignment between research intent, study design, and outcomes.

✅ You don’t need all four components to proceed , but try to fill as much as you can. Leave a field blank if it’s not applicable.
PS: There is another format called P-E-O (Population, Exposure, Outcome) which is useful for non-comparative questions. We will discuss this during the workshop.

PICO Table 

PICO Component Description Prompting Question Your Entry
P Patient / Population / Problem How would I describe a group of patients similar to mine?
I Intervention / Prognostic Factor / Exposure Which main intervention, prognostic factor, or exposure am I considering?
C (Optional) Comparison What is the main alternative to compare with the intervention?
O Outcome What do I hope to measure, accomplish, or affect?

Step 5: Apply the FINER Criteria

You have a beautiful PICO. But is it a question worth asking? The FINER criteria are your reality check. Use this table to critically evaluate your research question. Score each criterion from 1 (very low) to 5 (very high).

Criteria Question to Ask Score (1-5) Justification / Your Notes
F – Feasible Can we actually do this? Do the data sources exist? Are the endpoints captured in routine practice? Can we enroll enough patients?
I – Interesting Is it interesting to anyone other than us? Will it change clinical practice? Will a payer or regulator care?
N – Novel Does it genuinely fill an evidence gap, or are we just replicating a study from Europe?
E – Ethical Is it appropriate to conduct this study? For observational studies, this primarily relates to patient privacy and data protection.
R – Relevant Will the answer to this question impact decisions? Will it support a market access submission, change a clinical guideline, or inform patient care?

Step 6: Define the SMART Research Objective

 You’ve clarified your evidence gap and framed it with PICO. Now, it’s time to articulate a SMART research objective, a clear, fundable, and measurable statement that will drive the entire project forward.

Section A: RWE Project Concept Summary

Use the table below to summarize your strategic concept before finalizing the SMART objective.

Component Guidance Your Entry
Project Title Concise and descriptive (e.g., not “gMG study”, but “Hospitalization trends in gMG patients treated with SC therapy in KSA”)
Target Country/Region Specific MENA market (e.g., Saudi Arabia, UAE, Egypt)
Primary Stakeholder Who needs this evidence? (e.g., SFDA, DOH, payer, HTA)
Evidence Gap Brief summary of the gap this study will address
SMART Research Objective Will be developed in Section B
Proposed Study Design Retrospective cohort, registry analysis, payer survey, etc.
Potential Data Sources Likely sources (e.g., hospital EHR, insurance claims, national registries)
Initial Feasibility Check Short note on practicality (based on FINER test, availability of sites/data)

Section B: SMART Research Objective Builder

Use the SMART acronym to refine your study objective into one that is clear and actionable.

SMART Component Prompting Question Your Entry
S – Specific Does the objective clearly define the patient group, intervention/exposure, and setting?
M – Measurable Can outcomes be measured from the data source? What are the specific metrics?
A – Actionable Will the findings influence decision-making or fill a known gap?
R – Realistic Is it feasible within MENA constraints (data, access, sample size)?
T – Time-bound Is the timeframe for the study clearly defined (e.g., data from which years, follow-up period)?

✅ Final SMART Objective (Fill after completing the table above)

Write a one-sentence objective below. This is a statement, not a question (avoid “Does…?” or “Is…?”). This will guide protocol development.

Example: To assess the 12-month hospitalization rates among AQP4+ NMOSD patients in Saudi Arabia treated with eculizumab, using EHR data from two tertiary hospitals collected between 2020 and 2024.


Step 7: Match Your SMART Objective to the Right Study Design

You now have a SMART research objective. Step 7 ensures you select the most appropriate study design, aligned with your aim, data availability, and regional expectations , especially in MENA, where local data is emphasized.

Section A: Research Aim → Study Design Decision Matrix

Research Aim Common RWE Designs Typical Data Source Key Data Captured Alexion Example
Treatment Patterns Retrospective Cohort Study EHR, claims data Prescription fills, administration records, therapy dates gMG treatment sequence in UAE
Effectiveness / Comparative Retrospective or Prospective Cohort Study
Historical Controls (if needed)		 EHR, registries, trial arms Relapse rates, clinical scores, imaging, survival NMOSD relapse comparison on Eculizumab vs. SOC
Safety / Signal Detection Case-Control Study, Signal Mining PV databases, claims AE codes, lab alerts, co-medications Rare AE detection in long-term users
QoL / Economic Burden Cross-Sectional Survey, PRO + Claims Linkage PRO tools (EQ-5D, FACIT), payer data QoL scores, resource use, indirect costs HPP caregiver burden survey
Epidemiology / Burden of Illness Descriptive Registry, Cross-sectional study National registries, hospital discharge logs ICD-10 codes, demographics, prevalence PNH prevalence in KSA via registry
Clinical Pragmatism / Access Pragmatic Clinical Trial, Registry-nested Trial Hospital-based records + minimal trial data Routine practice outcomes, limited exclusion criteria (Hypothetical) SC formulation vs IV in real-life gMG patients
Comparative Realism in MENA Historical Control Study (FDA-accepted) BUT MENA requires local dataset Past trial arm data + local EHR population Efficacy from prior studies + regional confirmation Comparing new therapy to historical SOC using hybrid local + global

⚠ Note: While FDA may accept pragmatic trials and historical control designs, MENA agencies (e.g., SFDA, DOH, MOH Egypt) often expect local, recent, and population-specific data.

Section B: How to Choose, Guided Questions

Decision Question Your Notes
What is the main research aim (e.g., treatment pattern vs effectiveness)?
Is this Launch or Pipeline? Refer to Step 1.
What data source(s) are available locally? (hospital, payer, registry)
Is it ethical or feasible to randomize in this disease area?
Can historical data be used as a comparator? (Justify)
Would regulators or payers accept this study in our region?

✅ Final Output for This Step

Based on the table and reflection above, what study design are you selecting?

Proposed RWE Study Design
Rationale (one line summary)

Step 8: Ensure Statistical Rigor Through the Right Complexity Level

All RWE must meet statistical rigor, but not all studies require the same level of complexity. In this step, you’ll align your analysis methods to the decision being supported, the study design, and the feasibility of your data.

Reminder: “Complexity ≠ Quality”. It’s about matching the method to the decision.

Section A: Define the Decision Context

What is the primary purpose of this evidence? Tick one. This will guide the expected complexity of the analysis.

Decision Type Description Tick One
Regulatory Decision Label expansion, new indication, or safety profile change. Requires highest complexity.
Clinical Guideline Decision Influence expert consensus or published pathways. Moderate-to-high complexity needed.
Internal Strategy Decision For descriptive insight or local burden mapping. Lower complexity acceptable.

Section B: Match Study to Statistical Complexity Level

Complexity Level When to Use It Example Methods
High Complexity Comparative effectiveness, causal inference, confounding adjustment Propensity Score Matching, IPTW, Instrumental Variables, sensitivity testing
Moderate Complexity Association studies or time-to-event outcomes Logistic regression, Cox models, Linear regression, GEE, mixed-effects
Low Complexity Descriptive or exploratory studies Frequencies, means, medians, proportions, visual trends

✅ Note: Low complexity ≠ low quality. It just means the objective doesn’t require causal or comparative modeling.

Section C: Key Statistical Considerations 

Don’t ignore missing data! It’s not just about the percentage of missing data; it’s about the trend or pattern of what’s missing. A plan is mandatory.
Element Guidance Notes
Time-to-Event Analysis Use Kaplan-Meier curves, report medians and CI. Consider log-rank or Cox models.
Longitudinal Data Use mixed-effects models to manage repeated measures (e.g., over multiple clinic visits).
Missing Data Pre-specify approach (e.g., multiple imputation); run sensitivity analyses.
Small Sample Sizes Use exact tests (e.g., Fisher), or Bayesian inference. Justify the power limitations.

✅ Final Output for Step 8

















Step 9: Understand MENA Regulatory Frameworks

A perfect study design means nothing if regulators don’t buy into it. MENA is not uniform, you must understand and align with local expectations before you start.

⚠ “In MENA, regulatory success = scientific merit × localization.”

Section A: Country-Level Requirements Snapshot

Country / Region RWE Acceptance Local Nuance Submission Tip
Saudi Arabia (SFDA) High & Formalizing Local data preferred. PASS designs common. Use scientific advice. Position study as post-marketing.
UAE (MOH / DOH Abu Dhabi) Moderate to High GCC data accepted; UAE-specific better. Strong HTA interest. Follow formal advice; cite DOH RWE guidance.
Egypt (EDA) Emerging Local PI required. Cost-effectiveness focus. Go through CROs or academia. Submit to special study board.
Other GCC (Qatar, Kuwait) Moderate & Following GCC-level data fine. Emphasis on value to system. Submit as HTA evidence. Emphasize local relevance.

Section B: The “ADAPT” Strategy

In the absence of unified MENA regulation, follow this smart 3-step playbook:

  1. Adopt Global Standards: Build on FDA, EMA, and ISPOR guidance to gain baseline credibility.
  2. Adapt to Local Context: Address specific local needs such as:
    • Arabic-language PROs
    • Saudi PDPL & other regional privacy laws
    • Local IRB/ethics expectations and timelines
  3. Engage Early & Often: Never present a completed study as a surprise. Align with regulators and key stakeholders via pre-submission meetings and scientific advice.

✅ Output Box: Regulatory Readiness Checklist

Have you considered these key local requirements for your protocol?

Embedded in Protocol? Requirement Example

Local data / registry source e.g., SFDA or DOH evidence expectation

Local stakeholder alignment PI, ethics board, HTA reviewer

Arabic PROs if needed Especially for QoL endpoints

Privacy & consent aligned PDPL (KSA), Data Protection Law (UAE)

Early scientific advice planned Submitted inquiry to regulator before final protocol

Step 10: Conduct a Thorough Feasibility and Risk Assessment

Every great idea must pass a practical test: Can this be done? Feasibility is not about avoiding risk, it’s about identifying it early, scoring its impact, and having a realistic mitigation plan.

Part A: Site & Operational Feasibility Checklist

A full feasibility assessment is often completed by your clinical operations team or site partners using a detailed questionnaire. For our purposes, let’s focus on a high-level check.

📎 A comprehensive Feasibility Questionnaire Template is available for detailed site assessments. It covers site capacity, population estimates, regulatory timelines, staffing, and IRB/contracting logistics.

📝 If you’re completing this on your own for now, fill out the brief checklist below with your best estimates.

Core Feasibility Item Your Initial Assessment / Notes
Estimated number of eligible patients per site/per year
Ethics committee (IRB/EC) setup and estimated timeline
Site staff availability and experience (PI, coordinators)
Data capture methods (EHR, paper, PRO tools, labs)

Part B: Core Anchor 3, Risk Matrix Application

Use this matrix to list and score potential risks before initiating the project. The goal is to quantify, not eliminate. What can we proactively do to prevent or reduce the impact of this risk?

Risk Category Specific Risk Description Impact
(1–9)		 Likelihood
(1–9)		 Risk Score
(I × L)		 Mitigation Plan (Mitigate, Avoid, or Replace)
e.g., Recruitment Patient pool smaller than estimated 6 7 42 Adjust inclusion criteria. Engage referring centers.
e.g., Contracting Delay Long negotiation with academic hospital 5 8 40 Pre-engage with legal early. Have template ready.
e.g., Data Capture Site has no eCRF experience 4 5 20 Provide on-site training + support staff.












✅ Output Summary













Step 11: Finalize the RWE Synopsis

Summarize everything in one structured, strategic page. This synopsis is your internal go-to document, the bridge between concept, feasibility, and study activation.

📄 RWE Study Synopsis Template

Item Guiding Prompt Your Entry
Study Title Descriptive but clear
Research Objective SMART objective from Step 6
Pipeline or Launch Phase Tick one →
Primary Endpoint What are you measuring?
Target Country/Region e.g., KSA, UAE, Egypt
Target Stakeholder SFDA, DOH, payer, internal strategy, etc.
Study Design From Step 7 (e.g., retrospective cohort, survey, historical control)
Data Source(s) EHR, registry, PRO tools, etc.
Sample Size Estimate Approximate based on feasibility data
Site(s) Key partners or types of hospitals
Complexity Level Low / Moderate / High (Step 8)
Regulatory Fit Based on Step 9 — are you aligned with country guidance?
Top 3 Risks + Mitigations Short summary from Step 10
Next Action e.g., feasibility call, protocol draft, stakeholder alignment

✅ Final Output










>