Immune-modulating therapies, like bispecific T cell engagers (TCEs) and mRNA vaccines can be incredibly effective, but they can also trigger fast, hard-to-predict immune side effects that are costly if uncovered late in the development process. The question every team preparing for Investigational New Drug (IND) applications and first-in-human (FIH) decisions should be asking themselves is simple: How do we create an immunotoxicity strategy that is appropriate for IND submission and still executable on a real timeline?
It starts with a decision framework grounded in the ICH S8 “weight-of-evidence/cause for concern” mindset. This means starting with standard studies, then adding targeted work only when the data or mechanism of action (MoA) justifies it. The overall goal should always be to reduce avoidable delays (e.g., rework, last-minute study changes, misaligned endpoints) while also reducing human risk. To achieve this, drug developers and sponsors must be proactive and align on four important decisions.
Decision 1: Define Your Immune Risk Hypothesis
Start by writing a simple program-wide immune risk hypothesis. This plain-language document maps your therapy to the top immune safety risks you want to detect early, rule out, or monitor. Think of it as a hazard map the whole team can agree upon before key IND and FIH decisions are finalized. It is crucial to keep this hazard map clear and current, so review and revise it often. Three common hazards—cytokine release, immunosuppression, and organ inflammation—are usually enough to drive an actionable plan.
After that, conduct an early “human relevance” check so you’re not relying on a model that can’t answer the clinical question. For programs where immune activation is plausible, an in vitro cytokine release assessment using human whole blood or peripheral blood mononuclear cells (PBMCs) can be a practical early screen to inform what to watch in vivo and in the clinic.
Finally, make your plan executable. Schedule one cross-functional immune risk review before GLP toxicology testing starts and leave with two outputs: (1) your one-page hypothesis and (2) a short list of triggers that tell the team when to escalate. This early work can be an effective anchor for IND and FIH decisions that can help avoid unnecessary delays and stress.
Decision 2: Use Standard Toxicology as the Screen & Pre-Specify Triggers
Standard toxicity studies should be your first, consistent immunotoxicity screen, supported by a small set of endpoints that remain consistent. Achieving IND readiness means pairing that baseline with predetermined triggers, so the team isn’t tempted to improvise mid-study. A lean core battery of tests can usually be built from four elements:
- Clinical observations (including body temperature when feasible).
- Hematology with differential, emphasizing absolute counts of key white cell types.
- Clinical chemistry with a basic inflammation lens (e.g., globulins or albumin/globulin ratio where relevant).
- Gross pathology and histopathology of lymphoid organs (e.g., thymus, spleen, lymph nodes) and other target organs.
If consistent shifts in white cell subsets appear beyond pre-defined bounds, the trigger should be to add an immunophenotyping panel at the next scheduled sampling time point. This way, the program can confirm which immune populations are changing and whether the pattern is consistent with a meaningful immune effect. Similarly, if lymphoid organ histology suggests clear activation or depletion, the trigger should be to expand phenotyping or add one focused immune function readout to clarify whether the finding is likely to matter clinically rather than leaving it as an uninterpreted pathology observation.
Plan logistics up front to make triggers actionable. Modest amounts of extra blood and a small set of pre-labeled “hold” tubes can prevent the classic failure of seeing a signal but missing the chance to follow it. Proactive planning like this protects sampling windows and timelines.
Decision 3: Design Translatable Immune Monitoring
This decision helps carry a simple immune monitoring plan from nonclinical studies into FIH with minimal reinvention, using peripheral blood as the common bridge. ICH S8 specifically notes peripheral blood as useful for clinical bridging and recommends using absolute numbers of lymphocyte subsets in addition to percentages when evaluating treatment-related changes.
Keep the readouts succinct and easy to explain to scientific and non-scientific audiences:
- Report absolute counts + percentages.
- Focus on 3-4 cell types that map to your main risks: T cells (CD4/CD8 backbone), B cells, NK cells, and one myeloid bucket (monocytes or neutrophils, depending on mechanism).
- If cytokine release is a concern, add a small cytokine panel as a secondary layer—proportional to the risk hypothesis, not an open-ended exploration.
Finally, any shifts should be interpretable and actionable. It’s often best to standardize processes across studies, so trends mean something. The same methods and reporting format help reduce site-to-site noise and make unanticipated changes clear.
Decision 4: Pre-Plan the ‘If Positive’ Path
Positive immune signals are rarely self-explanatory, and the right follow-up turns uncertainty into a decision the clinical team can act on. This means proactively agreeing on courses of action that are only used when the data suggest a legitimate concern. Each course of action should be mapped to a fundamental question: Is the change meaningful? How severe is it? Can it be reversed?
Start with one functional tool for scenarios when an unexpected finding occurs. ICH S8 recommends an immune function study such as a T-cell dependent antibody response (TDAR) when additional immunotoxicity evaluation is needed, and a specific affected target isn’t identified. If a specific cell type is identified, targeted functional assays for that cell type may be appropriate. Put simply: if phenotype alone doesn’t tell you whether immune function is compromised, use a functional readout to support a clearer interpretation of risk.
Next, build reversibility into your plan. A recovery period (when warranted by signals) helps determine whether immune effects normalize and can directly influence how conservative FIH monitoring and escalation should be. Keep deeper mechanistic workups in reserve as optional, signal-driven tools, to be used only when they have the potential to change a decision.
Finally, have TDAR and any likely targeted assays readily available, budget time for triggered work, and predetermine who can authorize add-ons within 24–48 hours, so you don’t miss a critical sampling window.
A Final Word
New thinking around immunotoxicity doesn’t always mean maximal testing. It often means making some early, high-leverage decisions that protect human relevance, interpretability, and the development timeline under a weight-of-evidence mindset. When these four decisions are locked in before IND, unexpected findings become structured decisions rather than last-minute surprises that introduce risk, increase cost, and add delays.
