Pathways to Precision: Guideline for Programmed Death Ligand-1 and Tumor Mutation Burden Testing to Support the Selection of Immune Checkpoint Therapies in Lung Cancer Open Access

Immune checkpoint therapies have been shown in numerous clinical trials to have significant clinical benefits for lung cancer patients, and there are increasing numbers of indications and therapies available.^1–13  In lung cancer, pathologists have played a critical role in identifying patients who may respond best to these therapies. Initially, this was achieved using programmed death ligand-1 (PD-L1) immunohistochemistry, and more recently, tumor mutation burden (TMB) has emerged as another biomarker to potentially predict patients’ responses to immunotherapy.¹⁴  These tests have grown increasingly complex, with multiple companion diagnostic tests (CDx) for different approved therapies and indications.¹⁵  Many laboratories face challenging questions on the best practices for implementing and reporting PD-L1 and other predictive markers, such as TMB, to effectively guide treatment decisions. In this issue, a collaborative group from the College of American Pathologists, the Association for Molecular Pathology, the International Association for the Study of Lung Cancer, the Pulmonary Pathology Society, and the LUNGevity Foundation release a comprehensive guideline on PD-L1 and TMB testing in lung cancer.¹⁶ 

Also see p. 757.

After an extensive systematic review of the literature and the use of a comprehensive tool for grading, the group has developed 6 evidence-based recommendations for the use of PD-L1 and TMB testing in lung cancer to direct the selection of patients for immunotherapy. The recommendations, summarized in the Figure, chart a pathway for pathologists and pathology laboratories to support the selection of patients. The recommendations first recognize and support the need for an integrated analysis of PD-L1 and other targetable genomic biomarkers, such as ALK and EGFR, which have well-defined targeted therapies available. There is a correlation between PD-L1 tumor proportion score (TPS) and patient response and survival following immunotherapy.¹⁷  However, multiple studies have demonstrated a lack of benefit for patients with driver alterations in EGFR and ALK, and studies have demonstrated that combined therapy with EGFR and immune checkpoint inhibitors has high rates of interstitial pneumonitis.¹⁸  These findings underscore the need for comprehensive testing in lung cancer patients for both PD-L1 and molecular drivers. There is some debate on which patients should be tested, and this is likely a moving target as indications evolve and earlier-stage and neoadjuvant therapies are considered in patients.

Lung cancers are typically accessible and diagnosed by surgical resection, endobronchial ultrasound, or pleural fluid sampling. This can result in a number of specimen types that may have been fixed in formalin or other cytology fixatives. Further, lung cancer can spread to the bones, and these specimens can often require decalcification. There are benefits to testing on the best specimen available; however, pathologists must ensure that appropriate validation has been performed on all specimen types that are tested. This includes cases fixed with cytology fixatives or decalcification.¹⁹  Some studies have shown effects for some decalcification methods, and it is imperative to validate these assays under different use conditions before reporting.²⁰  In scenarios where multiple specimens are available, it can be challenging for pathologists to identify the best specimen for PD-L1 testing. There is some variability, but in general studies show concordance between small biopsies and cytology with resection samples for PD-L1^21,22 ; however, in limited or borderline biopsies composed of predominantly lepidic carcinoma, consideration for repeat testing is given by many pathologists.

The landscape of PD-L1 testing is rapidly evolving, with newer drugs, tests, and indications for testing. Most drugs have a companion diagnostic assay with a specific clone of PD-L1 that is linked to the use of a specific drug and indication(s). Although 3 of the 4 commonly used PD-L1 assays (22C3, 28-8, and SP263) demonstrate similar staining, the fourth assay (SP142) has been shown to have weaker staining.^23,24  The guideline suggests that when feasible, pathologists should use clinically validated CDx assays as intended based on data from the clinical trials that support the various therapeutic indications. The guideline suggests that laboratory-developed tests can be considered, but they should be validated according to the requirements of the local accrediting body.

In lung cancer, the TPS is used for PD-L1 assessment to enumerate the percentage of positive tumor cells. Clinical studies have demonstrated improved response rates for patients with higher levels of PD-L1 expression, especially in patients with a PD-L1 TPS greater than 50%.²⁵  Typically, TPS scores are grouped into negative (<1%), weakly positive (1%–49%), and strongly positive (≥50%) groups. The guideline suggests that pathologists should report the TPS percentage for cases but recognize that using ranges (often in 5%–10% ranges) may increase efficiency and decrease interobserver variability in scoring. It is critical that the ranges used map back to clinically relevant cut points to ensure that relevant clinical trial data can be used to guide decision making.

TMB testing has emerged as another biomarker and was shown in some early studies to predict the response to immune checkpoint therapy.¹¹  However, there are overall limited data to support the widespread use of TMB as an independent predictor of response to immune checkpoint inhibitors in lung cancer. Further, there can be significant technical challenges and costs associated with TMB testing.²⁶  In agreement with these points, the guideline highlights that there is insufficient evidence to support the use of TMB alone in patients with advanced lung cancer.

The guideline underscores the criticality of patient-centered care in pathology. By ensuring accurate, clinically validated, and timely biomarker testing, pathologists directly influence treatment decisions, and consequently patient outcomes. The immunotherapy space is rapidly evolving, and these guidelines and testing will need to continue to evolve to ensure that we as pathologists can continue to meet the needs of our patients to ensure optimal selection for these therapies.