Target Identification and Validation

## What Makes a Good Drug Target

A drug target is a molecule in the body whose activity can be modified by a drug to produce a therapeutic effect. The ideal target has strong causal evidence linking it to disease, is druggable (accessible to small molecules or biologics), has a favorable safety profile when modulated, and offers measurable biomarkers for clinical development.

Historically, many drugs were discovered empirically before their targets were known. Modern drug discovery inverts this by starting with target biology and designing molecules to modulate it.

## Target Identification Methods

**Genomics and GWAS**: Genome-wide association studies identify genetic variants associated with disease risk. If a variant affects a specific gene's function and associates with disease, that gene product becomes a candidate target. PCSK9 inhibitors exemplify this -- individuals with PCSK9 loss-of-function mutations have very low LDL cholesterol and reduced cardiovascular risk, validating PCSK9 as a target.

**Transcriptomics and proteomics**: Comparing gene expression or protein levels between diseased and healthy tissue identifies differentially expressed molecules. Single-cell RNA sequencing now enables cell-type-specific target discovery.

**CRISPR screens**: Genome-wide CRISPR knockout or interference screens in disease-relevant cell models systematically identify genes whose loss affects disease phenotypes. Pooled CRISPR screens can evaluate 20,000+ genes in a single experiment.

**Phenotypic screening**: Testing compounds in disease-relevant cellular assays without a predetermined target. Hits are then deconvoluted to identify the molecular target. This approach captures targets that reductionist methods might miss.

## Target Validation Approaches

**Genetic validation** provides the strongest evidence. Mendelian randomization (MR) uses genetic variants as instruments to estimate causal effects of target modulation on disease outcomes, essentially conducting a natural experiment. Targets with human genetic support have 2-3x higher clinical trial success rates.

**Pharmacological validation** uses selective tool compounds to demonstrate that modulating the target produces the expected phenotype in cell and animal models. Tool compound quality is critical -- poor selectivity leads to misleading results.

**Animal models**: Knockout/knock-in mice, disease models (EAE for MS, collagen-induced arthritis for RA), and patient-derived xenografts (oncology) test target hypotheses in vivo. Species differences in target biology and immune function limit translatability.

**Clinical biomarkers**: Demonstrating that target engagement correlates with clinical improvement in early human studies provides the most direct validation. PET ligands, pharmacodynamic biomarkers (e.g., LDL reduction for PCSK9), and circulating protein levels serve this purpose.

## Common Pitfalls

Target identification failures often stem from correlation-causation confusion (differentially expressed does not mean causal), poor animal model translatability, publication bias favoring positive results, and inadequate assessment of target safety (on-target toxicity from broadly expressed targets). Rigorous validation using multiple orthogonal methods reduces these risks.

## Key Takeaways

- Human genetic evidence (GWAS, Mendelian randomization) doubles clinical success probability
- CRISPR screens enable systematic genome-wide target identification in disease models
- Multiple orthogonal validation methods are essential to de-risk target selection
- Phenotypic screening captures complex biology missed by target-centric approaches