PET/CT-Guided Discontinuation of Immunotherapy in Metastatic Lung Cancer: A Paradigm Shift Toward Precision De-escalation

Introduction: The Dilemma of Indefinite Immunotherapy

The advent of immune checkpoint blockade (ICB) has fundamentally transformed the prognosis for patients with advanced non-small cell lung cancer (NSCLC). However, as we witness an increasing cohort of long-term responders, a critical clinical question remains unanswered: How long should treatment continue? Current guidelines often default to indefinite treatment or a fixed two-year duration, yet neither approach is rooted in robust biological evidence. Indefinite treatment not only increases the risk of cumulative immune-related adverse events (irAEs) but also places a staggering financial burden on healthcare systems worldwide.

Recent data from the National Network Genomic Medicine Lung Cancer Germany (nNGM) suggests that we may be over-treating a significant portion of our patients. By utilizing PET/CT as a functional imaging surrogate for deep molecular response, clinicians may be able to safely discontinue therapy, potentially improving survival outcomes through more vigilant monitoring and earlier intervention for resistance or new primary malignancies.

Highlights of the nNGM Analysis

• PET/CT-guided discontinuation of ICB was associated with a 65% reduction in the risk of death compared to continued treatment (HR 0.35; p = 0.002).
• Patients in the discontinuation cohort received an average of 18 fewer months of therapy, significantly reducing toxicity risks and costs.
• Systematic re-biopsy upon progression revealed that 28% of cases were not actually resistant disease, but second primary lung cancers (SPLC) requiring different management.
• Most progression events following discontinuation were successfully managed with local ablative therapies rather than systemic chemotherapy.

Study Design and Patient Population

This retrospective cohort study analyzed 455 patients across 21 nNGM centers who achieved at least two years of disease control on first-line ICB-based regimens. The study population was divided into two distinct groups: Cohort A (n=126), where treatment was discontinued based on PET/CT guidance, and Cohort B (n=329), where patients continued ICB without a structured PET/CT-guided stopping protocol.

The primary endpoint was overall survival (OS). Secondary endpoints included treatment duration, the nature of progression events, and a deep-dive into the tumor microenvironment via comprehensive genomic profiling, tumor-infiltrating lymphocyte (TIL) quantification, and spatial transcriptomics in patients who experienced persistent or progressive disease.

Key Findings: Survival and Treatment Duration

The results were striking. After a median follow-up of 55 months, the median overall survival for the PET/CT-guided discontinuation group (Cohort A) had not yet been reached. In contrast, the group that continued treatment (Cohort B) had a median OS of 82 months. The hazard ratio of 0.35 suggests a profound survival benefit for the structured discontinuation strategy.

Perhaps more importantly, this survival benefit was achieved with significantly less exposure to drugs. The median treatment duration was 27 months in Cohort A versus 45 months in Cohort B. This suggests that the biological benefits of ICB—likely mediated by long-lived memory T-cells—persist long after the drug is cleared from the system, provided a deep functional response has been achieved.

The SPLC Revelation: Not All Progression is Resistance

One of the most clinically impactful findings of the nNGM study was the result of systematic re-biopsies. In Cohort A, 28% of patients who appeared to have progressive disease were actually diagnosed with a Second Primary Lung Cancer (SPLC). This distinction is vital; while acquired resistance to ICB often necessitates a change in systemic therapy, an SPLC can often be treated with curative intent using surgery or stereotactic body radiotherapy (SBRT).

Multi-modal profiling revealed that these SPLCs had distinct biological profiles compared to the original metastatic lesions. They typically displayed features of primary resistance, such as low PD-L1 expression, low tumor mutational burden (TMB), and “immunologically cold” microenvironments. Identifying these cases early through a structured monitoring program allowed for more effective, localized clinical interventions.

Managing Progression: Local vs. Systemic Intervention

The study also highlighted a shift in how progression is managed in long-term responders. In the PET/CT-guided group, 53% of progression events were managed exclusively with local ablative treatments. In the continued treatment group, this figure was only 17%. This suggests that by stopping systemic therapy and monitoring closely, clinicians are better positioned to catch oligoprogressive disease that is amenable to local control, thereby extending the window of survival without the need for toxic salvage chemotherapy.

Expert Commentary: Biological Plausibility and Clinical Implications

The superiority of the discontinuation cohort may seem counterintuitive to those accustomed to the “more is better” philosophy of oncology. However, from an immunological perspective, it makes sense. Chronic exposure to ICB can lead to T-cell exhaustion, where the very cells meant to fight the tumor become dysfunctional due to overstimulation. A treatment-free interval may allow the immune system to maintain a more robust memory profile.

Furthermore, the high incidence of SPLC underscores the need for a “biopsy-first” mentality when long-term responders show new lesions. Clinicians should be wary of assuming that every new nodule is a failure of the original immunotherapy. The nNGM data suggests that PET/CT is an excellent tool for identifying which patients have reached a state of “metabolic dormancy” where treatment is no longer adding value.

However, it is important to acknowledge the limitations of this study. As a retrospective analysis, there is a risk of selection bias—patients who were chosen for discontinuation may have had more favorable baseline characteristics. While the researchers used robust statistical methods to account for this, the definitive answer will come from the ongoing prospective non-inferiority randomized trial within the nNGM network.

Conclusion: Toward a New Standard of Care

The nNGM study provides a compelling argument for a structured, PET/CT-guided discontinuation strategy in long-term responders with metastatic lung cancer. This approach not only safeguards patients from unnecessary toxicity and reduces healthcare costs but may also offer a survival advantage by facilitating the early detection of second primaries and oligoprogression.

For now, clinicians should consider PET/CT as a valuable tool in the decision-making process for patients who have completed two years of ICB. If a patient is in a complete metabolic response, a discussion regarding the safety and potential benefits of discontinuation is not only appropriate but evidence-based.

References

1. Frost N, Joosten M, Franzen J, et al. PET/CT-Guided Management of Immune Checkpoint Blockade and Post-Treatment Multi-Modal Profiling in Long-Term Responders with Metastatic Lung Cancer in the National Network Genomic Medicine Lung Cancer Germany (nNGM). Ann Oncol. 2025;S0923-7534(25)06329-X.
2. Hellmann MD, et al. Duration of Response to Pembrolizumab in Patients with Advanced NSCLC. N Engl J Med. 2020.
3. Garon EB, et al. Five-Year Overall Survival for Patients With Advanced NSCLC Treated With Pembrolizumab: KEYNOTE-001. J Clin Oncol. 2019.