Diagnostics & Personalized Medicine

Molecular Diagnostics: PCR, NGS, and Beyond

PCR-based diagnostics are the clinical workhorse. Viral PCR tests detect specific viral nucleic acid sequences — the standard for COVID-19, HIV, hepatitis viruses. Cancer PCR tests detect mutations characteristic of specific cancers (BRAF V600E in melanoma, EGFR mutations in lung cancer). PCR diagnostics are typically rapid, highly sensitive, and increasingly affordable.

NGS-based diagnostics offer broader analysis: whole genome or exome sequencing for rare genetic disease diagnosis, targeted gene panels for cancer or cardiovascular conditions, liquid biopsies (sequencing cell-free DNA from blood to detect cancer mutations without tissue biopsy), non-invasive prenatal testing (NIPT) (sequencing fetal DNA from maternal blood, largely replacing more invasive prenatal testing), and pathogen genomic surveillance for tracking variants.

Emerging platforms: CRISPR-based diagnostics (SHERLOCK, DETECTR) detect specific nucleic acid sequences rapidly in point-of-care settings. Isothermal amplification methods (LAMP, RPA) work without thermal cycling, allowing simpler instrumentation. Digital PCR enables ultra-sensitive detection of rare sequences — useful for minimal residual disease in cancer.

Biomarkers and Companion Diagnostics

Biomarkers are measurable indicators that predict disease, drug response, or outcomes. Diagnostic biomarkers identify whether disease is present (cardiac troponin for heart attack). Prognostic biomarkers predict disease course (BRCA mutations for breast cancer recurrence). Predictive biomarkers indicate whether a specific treatment is likely to work.

Companion diagnostics are tests paired with specific drugs. The FDA increasingly approves drug-diagnostic combinations together: HER2 testing + Herceptin (the original, 1998), EGFR mutation testing + EGFR inhibitors for lung cancer, BRAF V600E testing + vemurafenib for melanoma, PD-L1 testing + checkpoint inhibitors across cancer types, BRCA1/2 testing + PARP inhibitors for breast and ovarian cancer, CFTR mutation testing + tezacaftor/ivacaftor for cystic fibrosis.

The companion diagnostic framework improves drug efficacy (used in selected patients), justifies costs (treating defined responder populations), and excludes patients for whom a treatment probably wouldn't work. Companies like Foundation Medicine, Guardant Health, Tempus, and Natera specialize in this market, which exceeds $20 billion annually.

Genomic Medicine and Polygenic Risk

Polygenic risk scores (PRS) combine small effects of many common genetic variants to predict disease risk. A PRS for coronary artery disease might combine information from hundreds of thousands of SNPs. Applications include risk stratification for screening (identifying high-risk patients for intensive early screening), treatment selection, and — controversially — embryo selection in IVF.

PRS limitations: Most genomic studies have been conducted in populations of European ancestry, so PRS developed from these studies often perform much worse in other populations. Even the best PRS explains only a portion of disease risk, and knowing your score doesn't always translate clearly into clinical action.

Pharmacogenomics — using genetic variants to predict drug metabolism and response — is more clinically established: CYP2C19 variants affect clopidogrel efficacy, CYP2D6 variants affect dosing for dozens of drugs, and TPMT variants predict risk from thiopurine chemotherapy. Several major health systems now screen patients for pharmacogenomic variants at hospital admission.

Consumer and Wearable Diagnostics

Direct-to-consumer (DTC) genetic testing from 23andMe, AncestryDNA, Color, and others offers ancestry testing, health predisposition screening, and pharmacogenomics without a doctor's order. Clinical value is contested — major medical genetics organizations recommend testing in clinical contexts with genetic counseling. DTC tests can miss clinically important variants, generate false positives, and provide poorly contextualized information.

Wearable biosensors: continuous glucose monitors (CGMs) (Dexcom, FreeStyle Libre, Levels) now used broadly beyond diabetes management; heart rhythm monitoring through Apple Watch and competing devices; sleep tracking with increasing sophistication; continuous blood pressure monitoring (emerging). These detect events that wouldn't otherwise be found — silent atrial fibrillation, sleep apnea, glucose excursions in non-diabetics.

Microbiome testing — gut microbiome analyses at $99–$299 with personalized dietary recommendations — is growing despite the science being in early stages. Actionable personalized recommendations based on individual microbiome profiles remain largely speculative. The diagnostic frontier increasingly blurs lines between traditional medical testing, consumer products, and research data collection.

Wait, Actually…

The most common DNA test you'll ever take, you took at birth — and you probably don't know it.

In every US state, newborn screening is mandatory. Within the first 48 hours of life, blood is collected from a heel prick and tested for 30–60+ genetic and metabolic conditions. This program saves thousands of lives every year. Conditions like phenylketonuria (PKU), congenital hypothyroidism, sickle cell disease, and severe combined immunodeficiency (SCID) can cause severe disability or death within months of birth. With early detection, many can be effectively treated.

Worth pausing on: mandatory genetic screening of every person at birth, with long-term retention of biological samples in state laboratories, with research use that may not require parental consent. This is one of the largest population genetic surveillance programs in human history. It exists for good reasons — but it represents a remarkable expansion of state biological surveillance that has happened with almost no public discussion. If you're researching genetic privacy (BP7) or public health law (BP6), newborn screening is one of the most consequential and least examined examples of how genetic information flows in the US health system.

---