Beyond Drug Choice: Optimizing Testing, Sequencing, and Care Pathways for Immunotherapy in Advanced NSCLC

Key Takeaways When tissue availability is limited or delayed, plasma testing for biomarkers can be a practical bridge for treatment decisions. Initiation of immunotherapy too early can complicate targeted therapy later in the treatment sequence and, in some cases, amplify toxicity risk. Education on immune-related adverse events, clear escalation pathways, and appropriate supportive care are core components of safe immunotherapy delivery. When deciding on immunotherapy in advanced non-small-cell lung cancer (NSCLC) , it is tempting to focus on questions like, “Which PD-(L)1 agent is best?”, “Which chemotherapy backbone is most appropriate?”, and “What is the best clinical trial?” However, perhaps a more profound question is, “What is the ideal process for NSCLC management from diagnosis to clinical benefit?” In everyday practice, patients do better when testing, sequencing, toxicity management, and access barriers are treated with the same rigor as tumor biology. Jamie Chaft, MD: Biomarker Testing: Not a Checkbox But a Safety Intervention One uncomfortable truth in thoracic oncology is that many issues with immunotherapy start before the first infusion. Biomarker testing is often framed as an efficacy tool for choosing the best drug. In reality, biomarker testing is equally a safety tool for prescribing the right drug at the right time to the right patient. In the field, biomarker testing turnaround time should be treated as a modifiable variable, not a fixed constraint. If a clinic routinely receives actionable results after treatment has already started, it is indicative of a workflow problem that needs to be addressed. One practical strategy for this problem is plasma-based testing. Tissue remains foundational, but real-world barriers like insufficient tissue, delayed procedures, and slow processing are common. Plasma testing can shorten the time that results are received and reduce the odds that a patient is started on therapy blindly due to a ‘call to action.’ The goal is not to replace tissue biopsy but rather to not hinder clinical decision-making because of logistical constraints. Clinical Pearl: Build a Day 0 order set for advanced NSCLC that automatically triggers comprehensive molecular profiling that includes PD-L1. Pair it with a parallel plan for inadequate or delayed tissue (plasma testing, rebiopsy, and clear timelines for escalation). Ticiana Leal, MD: When NGS Is Delayed: Build an Algorithm Before a Crisis It is not uncommon to encounter an older patient with comorbid conditions who lives far from the clinic and anxiously awaits their results, particularly in rural settings. There is not 1 universally correct response. Every clinic should define, in advance, what constitutes an acceptable waiting period and implement a well-prepared plan when waiting becomes unsafe or impractical. This is a subtle but crucial distinction. The question is not simply, “Should we wait?” A better question is, “What is our standardized approach when waiting becomes clinically risky or operationally impossible?” That plan must incorporate disease progression, symptom burden, and the patient’s ability to return to the clinic quickly if new information demands a pivot. Clinical Pearl: Create an explicit pathway that accounts for: Disease tempo and symptom burden (ie, what triggers urgent treatment) Factors that drive immunotherapy initiation before testing results are received A bridging strategy that remains pivotable when molecular results are ready A closed-loop process to ensure results are reviewed and acted on immediately Jamie Chaft, MD: PD-L1 Testing Is Necessary But Never Sufficient PD-L1 is often viewed as a single switch whereby high PD-L1 expression equals immunotherapy and low PD-L1 expression equals combination therapy. This is a helpful starting framework, but it can create false certainty. In practice, PD-L1 is 1 piece of the decision-making process but not the whole puzzle. Other factors such as patient fitness, comorbidities, disease burden, pace of progression, and molecular profile must also be considered. If PD-L1 expression is high but a targetable driver is also likely, the healthcare professional (HCP) should pause. Conversely, if PD-L1 expression is low, adding more agents may not be appropriate, especially if the patient’s physiology or support system cannot tolerate the related toxicities. One common theme that emerged from a series of workshops that we did for this educational program was that more drugs can mean increased toxicity without guaranteed incremental benefit. Intensification might be appropriate for a carefully selected, fit patient with bulky disease, particularly if there is a plan to reassess and de-escalate. HCPs should avoid aggressive or rapid treatment changes and instead take a careful, stepwise approach. The same principle applies to baseline risk assessment. Interstitial lung disease and autoimmune conditions are not minor comorbidities when immunotherapy is being considered; they can (and should) directly affect treatment choices. It is not practical to rely only on the radiology report. Imaging results should be reviewed, and if there is any concern of fibrosis, a radiology consult is necessary. This step can prevent cases of severe pneumonitis. Clinical Pearl: Before starting immunotherapy, document the following: Autoimmune history (including “quiet” inflammatory bowel disease or rheumatologic disease) Baseline pulmonary risk (interstitial lung disease/fibrosis, prior pneumonitis, oxygen needs) A symptom-reporting plan (who to call and what triggers urgent evaluation) Ticiana Leal, MD: Targetable Alterations: Sequencing Affects Toxicity, Not Just Efficacy One of the most high-stakes lessons in modern NSCLC management is that sequencing is not merely a matter of response rates but also safety. If a targetable alteration is plausible, an “immunotherapy-first” approach can create downstream risk. Although many driver-positive tumors benefit less from immunotherapy, treatment sequence can magnify toxicities, particularly when targeted therapy follows prior immunotherapy exposure. This is why biomarker delays are not simply burdensome but also dangerous if they push HCPs toward a decision that later becomes difficult to undo. In the workshops, the phrase “give it pause” emerged consistently. When molecular data are pending and driver mutations are likely, it is important to resist the urge to start immunotherapy simply because PD-L1 results are available. Clinical Pearl: If you must start treatment before receiving molecular results, choose approaches that preserve flexibility and avoid irreversible sequencing errors, particularly in patients with any meaningful chance of actionable drivers. Jamie E. Chaft, MD: Immune-Related Adverse Events: Education and Timing Are Key Immune-related adverse events are not uncommon, and they are not always obvious. For immunotherapy to be effective in the real world, toxicities must be recognized and managed early in the treatment pathway. A single operational change can be transformative: the infusion visit should not be the first time symptoms are discussed. Patients should be informed that reporting symptoms early is not bothersome, and staff must utilize a clear triage pathway that triggers evaluation and treatment quickly. Without such a system in place, toxicity becomes reactive, emergency-based care, but with this system, toxicity becomes manageable outpatient care. Standardizing response principles matters. Suspected pneumonitis requires rapid evaluation and timely steroid initiation when appropriate, with careful follow-up and taper. Higher-dose steroids often require attention to supportive measures, including prophylaxis when indicated. Diarrhea/colitis requires early contact, exclusion of infection, and avoidance of reflex self-treatment that can delay proper management, and severe toxicities demand escalation. In fact, some scenarios appropriately eliminate rechallenge. Clinical Pearl: Provide every patient with a 1-page “IO Toxicity Action Plan” on Day 1 that includes: Red-flag symptoms (respiratory, GI, and neurologic symptoms; severe fatigue/weakness) Exactly who to call (including after-hours instructions) A clinical triage algorithm that empowers nurses and APPs to act promptly Ticiana Leal, MD: Operationalizing Multidisciplinary Care in Real-world Clinical Practice Finally, individual patient access issues must be treated as a clinical reality, not a footnote. Travel distance, caregiver limitations, costs, and limited subspecialty availability shape what treatment is safe and feasible. When a formal tumor board is not an option, virtual multidisciplinary touchpoints can still promote essential functions like shared decision-making, early input, and coordinated sequencing. Equally important is the early involvement of patient navigators and social workers. Transportation resources, caregiver support, and communication preferences are not social issues; rather, they are part of the treatment plan because they determine whether patients can complete testing, attend visits, and report toxicity promptly Clinical Pearl: Make access planning routine. At the first visit, document transportation, caregiver support, and communication preferences, and connect patients to navigation resources before therapy begins. Final Thoughts Ultimately, the philosophy of high-quality NSCLC immunotherapy care is built on anticipation. Anticipate delays and build pathways. Anticipate sequencing consequences and protect future options. Anticipate immune toxicity and educate patients on the importance of reporting symptoms early , then empower your team to respond decisively. When such systems are in place, immunotherapy is not just administered; it is delivered safely, equitably, and with the greatest chance of durable benefit. Your Thoughts What process or protocol might you consider implementing or refining to improve the safe and effective use of immunotherapy in advanced NSCLC? -->