Need & Solution Specification Framework

NB: this is an old thinking framework trying to distill a lot of things I'd learned into a more actionable biotech project evaluation map. Has not been updated since ~2019. May be updated for how incoming trends in ML, automation, and regulatory changes affect this.

1. Need Specification
- a. Need Identity
  - i. Framework: Describe need @ 5 levels of understanding: grandma, undergrad in field, graduate student in field, professor near field, industry vet in field
  - ii. Need Components
    - 1. Clinical manifestations and differential diagnosis criteria
      - a. Symptoms and etiology
        
        i. Prodromes, heredity, and exposure risks
        
        ii. Symptoms that motivate patients to seek care
        
        iii. Disease progression characteristics
        
        iv. What we know about molecular-, cell-, organ- and tissue-level etiologies
        
        v. History of pathology
        
        Common patient histories
        
        History of medical model of understanding, detecting, and treating the condition
      - b. Patient populations affected
        
        i. Size, trends, locations, other characteristics
        
        ii. Heterogeneity of disease stage and symptoms @ diagnosis
        
        iii. Epidemiology
      - c. Differential diagnosis discriminatory criteria
        
        i. Inclusionary criteria and tests
        
        ii. Exclusionary criteria and tests
        
        iii. Common misdiagnoses
        
        iv. Is barbarian knowledge on diagnosis available from experienced doctors?
        
        v. Prognostic correlations
      - d. Setting / implementation of care
      - e. High-level trends and risks
        
        i. Risk factors in health
        
        ii. Policy
        
        iii. Economic
    - 1. Mechanisms
      - a. Molecular interactions – known + putative
      - b. Cellular populations involved
        
        i. Homeostatic function of each
        
        ii. Function in disease / pathology area for each
        
        iii. Response to inflammation for each
        
        iv. Known significant interactions
        
        v. How each would try to return to homeostasis
      - c. Tissues and organ systems affected
      - d. Patient heterogeneity
        
        i. GWAS, OMIM, and ClinVar results
        
        ii. Differences in disease presentation
        
        iii. Common co-morbidities
        
        iv. Environmental and lifestyle risks
        
        v. Disease state at the time of diagnosis
        
        vi. Side effect patterns by co-morbidity, polypharmacology, incidence in given backgrounds
        
        vii. Differences in response to current treatment
        
        viii. Access to care / clinical experience
        
        ix. Idiopathic forms of the disease
    - 1. Current treatments / solutions to need
      - a. Efficacy of current treatments
        
        i. Disease stabilization / progression / regression
        
        Long-term efficacy
        
        Long-term patient adherence
        
        ii. Side effect profiles
        
        iii. History of indications each treatment approved for
        
        iv. Interactions and effects in co-treatment
        
        v. Polypharmacology of treatment with emphasis on treatments used for co-morbidities
        
        vi. Criteria upon which previous / current solutions were granted market approval
        
        vii. Traditional medicine and supplements used for the condition
      - b. Mechanism of action of current treatments, including hypotheses when MOA has not been directly shown
      - c. IP ownership landscape
        
        i. Architecture of claims
        
        ii. Scope and nature of accepted inventive steps
        
        iii. Unprotected prior art
        
        iv. What "skilled in the art of" would mean for this particular field
        
        v. Patentability criteria of past, present, and emerging solutions if obtainable
      - d. Mfg capacity landscape
        
        i. cGMP requirements for the composition of matter
        
        ii. Identify any rare components / equipment needed
      - e. PBM metrics on solutions: scheduling, coverage, performance
  - iii. Need Ecosystem
    - 1. Affected parties
      - a. Patient characteristics
      - b. Patient group advocacy
      - c. Community health impacts
    - 1. Morbidity / mortality costs
    - 1. Costs to affected parties
    - 1. Adjacent opportunities currently infeasible
    - 1. Infrastructure
    - 1. Systems in which the need currently exists / blocks some solution
      - a. Academic research models, methods, etiology debates, diagnostics development / biomarker search
      - b. Start-ups attempting to solve same or similar problem
      - c. Attempts by larger biotech/pharma to solve problem
      - d. Clinical manifestation and management of the problem
      - e. Regulatory oversight of previous solutions / precedents
        
        i. Evolution of regulatory standards for mgmt. of solution precedents
      - f. BRICS and developing markets perspective
- b. Stakeholder Perspectives on Need, for each: motivations, fears, profits, losses, QALY calculations, cost-benefit considerations
  - i. Patients and families
  - ii. Doctors and clinical staff
  - iii. Insurers and pharmacy benefit managers
  - iv. Venture capitalists during solution precedents
  - v. Founders involved in previous attempts to solve
  - vi. Patient advocacy groups involved in research, lobbying, or clinical education
  - vii. Clinical educators
  - viii. Diagnostic test makers and lab types
  - ix. Biotech / pharma companies
  - x. FDA / EMA
  - xi. Other countries' regulatory bodies / health systems
- c. Need Precedents
  - i. Previous models of need – how was it recognized by:
    - 1. Clinicians
    - 1. Advocacy groups
    - 1. Scientists
    - 1. Pharma
    - 1. FDA, CDC
    - 1. Epidemiological trends
    - 1. Environmental risk factors
    - 1. Genetic risk factors
    - 1. Subpopulations where need is particularly prevalent
  - ii. Treatment advancement precedents in field of need
  - iii. Key scientific advancements in understanding / articulation / mechanism of need
  - iv. Technological progress realized in previous evolution of need recognition or treatment options / efficacy
  - v. Major revisions to diagnostic criteria
- d. Critical Gaps in Progress on Need
  - i. Etiological unknowns
  - ii. Druggability
  - iii. Technological development
  - iv. Clinical recognition / capacity
  - v. Commercialization barriers (mfg, capital, IP)
  - vi. Why regulatory bodies have rejected or approved previous solutions
- e. Need Ecosystem
  - i. Incentives, authorities, IP creation / licensure / transfer, KOL bodies (e.g., AACR) for and between each of the following:
    - 1. Clinicians
    - 1. Advocacy groups
    - 1. Scientists
    - 1. Pharma
    - 1. FDA, CDC
    - 1. Epidemiological trends
    - 1. Environmental risk factors
    - 1. Genetic risk factors
  - ii. Risks to and opportunities for each from a new standard / solid progress (e.g., your solution, those of competitors) in etiological understanding, treatment possibility, new infrastructure needs and capacities, diagnostic ability, or cost / side effect differentiator
  - iii. Strategic goals / institutional mandates of each
  - iv. Articulation of the kind of leverage each of the above tends to be pliable to (e.g., what things cause them to move resources or collaborate?)
- f. Biological Complexity
  - i. GO and PPI networks involved—take with a grain of salt
  - ii. Current scientific, clinical, and insurer models for accepted disease etiology @ symptomatic, molecular, diagnostic, prognostic, and treatment effects / side effect risk factors
  - iii. Alternative hypotheses for disease etiology at genetic, molecular, cellular, or tissue systems levels
  - iv. Articulate what kinds of left-field biological disruptors of the current models could be possible
  - v. Current literature of patient heterogeneity and genetic risk variants
  - vi. Common comorbidities for target disease / condition
  - vii. Variance in prognostic outlook and significant prognostic signals / efforts to find such prognostic signatures
  - viii. Involvement of the immune system
- g. Economics of Need
  - i. Patient costs
  - ii. Insurer costs
  - iii. Medical costs (how many specialists involved)
  - iv. Typical costs for precedents / prior solutions
  - v. Epidemiological impact
1. Solution Specification
- a. Describe the current solution(s) to the problem @ 5 levels of understanding: grandma, undergrad in field, graduate student in field, professor in field, industry vet in field
- b. Imagine and articulate the ideal, non-resource-constrained solution to the need
  - i. What would it consist of? Drugs, biomarkers, assays, production techniques, cell types, biopsies, co-treatments?
  - ii. How would it work?
  - iii. How would it be used?
  - iv. Which patients would it work for and how would they be identified or treated in the clinic?
  - v. How would it impact patients' lives?
  - vi. How would it impact doctors and clinical staff?
  - vii. How would it be made?
  - viii. How would it impact insurers or pharmacy benefit managers?
  - ix. What would regulators think about it? What aspects would they be most and least worried about?
  - x. How would potential competitors in pharma/biotech think of it?
  - xi. What additional benefits would the solution give?
- c. Describe the difference between Sections A and B
  - i. Barriers in biological understanding
  - ii. Barriers in technological capability
  - iii. Lack of specific datasets, methods, tests
- d. Work backwards to break section C into key changes needed to realize section A with key proofs for each, including replication across models with appropriate statistical power and analysis methods
  - i. Disease biology
  - ii. Diagnostic tools, methods, standards
  - iii. Technological breakthroughs: chemistry, biologics, cell therapy, gene editing, allogeneic sources, stem cell bio, delivery, sequencing, molecular probes, biologics mfg. etc
  - iv. Clinical practice
  - v. Regulatory oversight / new regulatory ground
- e. Current Status – update as work progresses
  - i. Preliminary data
    - 1. Put data together in a deck and update it as new data emerges
      - a. Hypothesis → methods → results → interpretation
        
        i. Include risks retired → alternative interpretations of data → new risks introduced
    - 1. How well current data accords with previous literature
    - 1. Internal hypothesis model of system and treatment effect
    - 1. Key risks retired: mechanistic, human sample, diverse human samples, immunologically-complete animal model, dose-exposure relationship, toxicology, PKPD, off-target effects
    - 1. Narrative arc of data: significance, innovation, overarching hypothesis, key findings, their meaning, and critical next questions
  - ii. Ongoing work
    - 1. Key proofs being obtained
    - 1. Work sites, capacities thereof
    - 1. Collaborators
  - iii. Intellectual property status
    - 1. Inventors and affiliations + motivations if necessary
    - 1. Current IP status: provisional, filed, perfected
    - 1. List of prior disclosures, even in confidential environment
    - 1. Freedom to operate analysis
    - 1. Patentability analysis
  - iv. Stakeholder engagement
    - 1. Key questions, findings, and perspectives from discussions / interviews with:
      - a. Patients / patient advocates
      - b. Doctors in the field
      - c. KOL scientists
      - d. Insurers (when available)
      - e. Regulators (when available)
        
        i. Seek perspectives of colleagues working on similar projects in EMA, Japan, BRICS
  - v. Articulate difference between E and B to generate milestones and events that help the company formulate and communicate strategy
    - 1. Organize by which differences most likely to be altered by technical progress
    - 1. Organize potential generations of the solution to be built
      - a. Minimal needed features for launch (clinical trials)
        
        i. How it needs to work
        
        Define critical quality attributes
        
        Define critical performance characteristics
        
        Potential side effects by severity, including etiology, detection method, and risk to patients
        
        ii. How it needs to be made / delivered
        
        Supply chain and mfg. / delivery environment
      - b. Value-adding parameters
        
        i. What those parameters consist of
        
        Clinical trials: biomarkers, companion diagnostics, non-invasive data sources
        
        Additional disease indications
        
        Potential combination treatments
        
        Infrastructure investment pay-offs
        
        Datasets acquired and interpreted
        
        ii. Who would find each parameter more valuable: partners, regulators, others
- f. Key Milestones
  - i. Data needed to file pivotal IP
  - ii. Data needed for seminal publications
  - iii. Data needed for strategic partnerships
  - iv. Data needed to file IND
  - v. Data needed to begin Phase 0/1
  - vi. Data needed for market approval
  - vii. Data / success needed for IPO / M&A
- g. TPP
  - i. Indications and usage (clinical target, molecular target, expected effects)
  - ii. Dosage and administration (PKPD profile needed, if estimable)
  - iii. Dosage forms and strengths ( + co-treatments)
  - iv. Contraindications (anticipated or measured; comorbidity risk factors)
  - v. Warnings and precautions (potential negative outcomes / interactions)
  - vi. Adverse reactions (anticipated range of side effects based on drug mechanism and expression of target in other cells / tissues)
  - vii. Drug interactions (hard to predict de novo)
  - viii. Use in specific populations (target population, additional patients that could benefit)
  - ix. Drug abuse and dependence
  - x. Overdosage (worst case overtreatment effects)
  - xi. Description
    - 1. Composition of matter
    - 1. Indicated use
    - 1. Clinical practice
    - 1. Formulation
    - 1. Adverse event / side effects monitoring
  - xii. Clinical pharmacology (PKPD, biodistribution, dose-exposure, excretion)
  - xiii. Nonclinical toxicology
1. Translational Plan
- a. Preclinical mechanistic studies
  - i. List of disease models, features of human disease it recapitulates, idiosyncratic risks it introduces for each
    - 1. Cell lines / iPSCs
    - 1. Organoids
    - 1. Primary explants
    - 1. Primary cell organoids
    - 1. GEMMs
    - 1. Xenotransplantation
    - 1. Chemical or genetic exposure
  - ii. Endpoints necessary to show MOA
    - 1. On-target efficacy
    - 1. Off-target effects
    - 1. Side effects
    - 1. Optimization opportunities
  - iii. Correlation ain't enough, knock something out!
  - iv. Necessity and sufficiency of molecular targets
  - v. Dose-exposure relationship
- b. IND prep
  - i. Disease models and effect endpoints
  - ii. Safety profile: ADMETox
  - iii. PKPD
  - iv. Drug metabolism
  - v. cGMP mfg pilot
  - vi. Biomarker / companion diagnostic development
    - 1. Determine how biomarkers / dx to be used and whether / when investigational device exemption needs to be applied for
- c. Regulatory engagement
  - i. Prev regulatory standards for comparable products
  - ii. Relevant industry guidances and press releases
  - iii. Pre-IND meeting planning
  - iv. Protocol development / change / update procedure
  - v. IRB oversight and composition planning
- d. Clinical trial planning
  - i. Site selection
  - ii. Protocol development
    - 1. Inclusion criteria
    - 1. Exclusion criteria
  - iii. Protocol management / implementation
  - iv. Chain of custody for samples
  - v. Endpoints to be run on samples
  - vi. Data & access / encryption mgmt. strategy
- e. Phase 0 or diagnostic sensitivity
- f. Phase 1 or diagnostic specificity
- g. Phase 2
- h. Phase 3
- i. Additional indications
- j. Full TPP, update at each stage
1. Business Plan
- a. Value Model
  - i. Significance statement
  - ii. Innovation statement
  - iii. Overarching hypothesis: if successful, then ________
  - iv. Value to patients
  - v. Value to doctors
  - vi. Value to clinical practice
  - vii. Value to pharmacy benefit managers / insurers
  - viii. Regulatory or professional body recognition of problem urgency, scope, impact, etc
  - ix. Market value
    - 1. Present market for drugs used currently
    - 1. Market segmentation
    - 1. Parameter space of all possible solutions to the need
    - 1. Competitor pipelines w/ analysis
    - 1. Future market trends
  - x. Market differentiation
    - 1. Features / fronts by which competition plays out
    - 1. Infrastructural features that enable market access
    - 1. Proactive regulatory engagement
    - 1. Treatment delivery / deployment
    - 1. Partnering strategy—hospitals, patient advocacy groups, collaborators, academic centers, regulatory bodies, etc.
  - xi. Intellectual property
    - 1. Freedom to operate analysis
    - 1. Patentability analysis
    - 1. Licensing considerations
- b. Project Model
- c. Need Model
  - i. Etiology Model: molecular, cell, tissue, organ, host, patient-to-patient variability
  - ii. Intervention Context Model (Clinical Perspectives)
    - 1. Diagnostic criteria and common misdiagnoses
    - 1. First-line treatments and efficacy monitoring
    - 1. Second-line treatments
    - 1. Monitoring of disease progression
    - 1. Clinical setting
    - 1. Healthcare professional education, training, experience
    - 1. Detection of clinical complications of treatment
  - iii. Intervention Model Mechanism of Action
    - 1. Drug MOA by cell type, genetic background, disease stage – can be hypothetical but should include what kinds of data can confirm or reject when possible
    - 1. On-target activity and effects
    - 1. Off-target activity and effects (may be putative)
    - 1. How treatment alters disease etiology, progression, co-morbidity
    - 1. Long-term usage
  - iv. Addressable Population Model
    - 1. Epidemiological trends by geographic area and environmental exposure
    - 1. Genetic risk variants
    - 1. Environmental risk variations
    - 1. Patient heterogeneity
      - a. Symptomology
      - b. Demographic
      - c. Comorbidity background
    - 1. Accessibility of health care
  - v. Alternative models and what new scientific data would validate them
- d. Customer Model
  - i. Patient
  - ii. Doctor
  - iii. Insurer
  - iv. PBM
  - v. Pharma
  - vi. FDA / EMA
  - vii. Patient advocate groups
- e. Competitor Model
  - i. Competitor landscape
    - 1. Pipelines
    - 1. Partnerships
    - 1. Star personnel & histories
    - 1. IP opportunities
    - 1. Funding and development stage
  - ii. Strategic vulnerabilities in ecosystem incumbents
    - 1. Pipeline failures
    - 1. M&A in the same area that didn't work out
    - 1. Areas that they have a lot of marketing, distribution, or manufacturing investments concentrated
- f. Regulatory Model
  - i. Precedents for regulatory engagement for similar projects
  - ii. Relevant regulatory guidances and interactions
  - iii. List of anticipated regulatory concerns—update after meetings
- g. Drivers & Sinks
  - i. Value Drivers
  - ii. Need Drivers
  - iii. Progress/Success Drivers
  - iv. Risk Drivers
  - v. Competition Drivers
- h. Present Value Justification
  - i. Data
  - ii. IP
  - iii. rNPV for products based on market trends
  - iv. Partnership goals
  - v. Pipeline
- i. Future Value Justification
  - i. Clinical data
  - ii. Pipeline development
  - iii. Scientific / medical leadership
  - iv. Partnership strategy
  - v. Exit options
- j. Risk Management Strategy
  - i. How could your solution fail?
  - ii. What can you do to detect early signs of a failure point?
  - iii. What's your plan for fixing them?
  - iv. Which risks get retired at which milestones and why
  - v. Which aspects of the solution, opportunity, or translation process are concretely-established medical science and which are less well-settled and subject to potential changes?
1. Emergent Opportunities (things that become possible as 1-4 progress)
- a. Synergies / Cantilevers – where meeting a given milestone significantly lowers the cost to entry of an additional value-creating opportunity
- b. Field Futurism – where the underlying scientific frontiers are headed and how to make win-wins out of project planning for current priorities and being able to realize emerging opportunities later (e.g., building specialized infrastructure that can find broader application after a given tech bottleneck is surpassed)
- c. Leftfield Biological Realities
- d. Collaborations and Partnerships
- e. Patient Advocacy
- f. Platformization
- g. Swarm Mode
- h. Human Benefits!
  - i. Mortality reduced
  - ii. Lost economic value restored
  - iii. Quality of life improvements
  - iv. What happens to health when your patents expire or your tech becomes cheap and easy?
  - v. Plan to actually interact with the human beings whose lives your efforts will touch—do not stay away from clinics, patient groups, or families
- i. Application to emerging markets
- j. Care Support
  - i. Biomarkers for diagnosis
  - ii. Biomarkers for treatment monitoring and follow-up
  - iii. Digital infrastructure
  - iv. Clinical IT integration and extension
  - v. Liability reduction (e.g., gene therapy monitoring for pay-for-performance models)