Molecular Testing for Treatment of Metastatic Non-Small Cell Lung Cancer: How to Implement Evidence-Based Recommendations

Health care providers face considerable challenges in implementing biomarker testing recommendations for diagnosis and management of non-small cell lung cancer (NSCLC). This report provides insights into key challenges and recommendations for molecular testing of patients with metastatic NSCLC, summarized from a multidisciplinary team of experts spanning academic, community, and integrated health systems.

Abstract

The recent discovery of relevant biomarkers has reshaped our approach to therapy selection for patients with non-small cell lung cancer. The unprecedented outcomes demonstrated with tyrosine kinase inhibitors in molecularly defined cohorts of patients has underscored the importance of genetic profiling in this disease. Despite published guidelines on biomarker testing, successful tumor genotyping faces significant hurdles at both academic and community-based practices. Oncologists are now faced with interpreting large-scale genomic data from multiple tumor types, possibly making it difficult to stay current with practice standards in lung cancer. In addition, physicians’ lack of time, resources, and face-to-face opportunities can interfere with the multidisciplinary approach that is essential to delivery of care. Finally, several challenges exist in optimizing the amount and quality of tissue for molecular testing. Recognizing the importance of biomarker testing, a series of advisory boards were recently convened to address these hurdles and clarify best practices. We reviewed these challenges and established recommendations to help optimize tissue acquisition, processing, and testing within the framework of a multidisciplinary approach.

Implications for Practice:

Although several professional societies have incorporated biomarker testing recommendations into clinical practice guidelines for the diagnosis and management of non-small cell lung cancer (NSCLC), health care providers still face considerable challenges when establishing and implementing these standards. Developing and instituting protocols to ensure that all appropriate patients are tested for molecular biomarkers requires communication among the various specialists involved in the care of patients with NSCLC. This report provides insights into key challenges and recommendations for molecular testing of patients with metastatic NSCLC, summarized from a multidisciplinary team of experts spanning academic, community, and integrated health systems.

Introduction

Lung cancer is the leading cause of cancer-related mortality worldwide, accounting for an estimated 27% of all cancer deaths in the U.S. during 2014 [1, 2]. Non-small cell lung cancer (NSCLC) composes approximately 84% of lung cancer cases [2] with more than half of NSCLC patients presenting with advanced-stage disease [3, 4]. Historically, patients with advanced-stage NSCLC have been treated with platinum doublet chemotherapy. Unfortunately, this approach results in only a modest gain in survival at the cost of considerable toxicity [5, 6]. Recently, a greater understanding of the molecular pathways driving carcinogenesis led to the development of novel agents in NSCLC. Patients with epidermal growth factor receptor-activating (EGFR-activating) mutations or anaplastic lymphoma kinase (ALK) rearrangements have demonstrated unprecedented outcomes when treated with a first-line tyrosine kinase inhibitor (TKI) [4, 7–13]. Recognizing the importance of a timely and accurate diagnosis—including tumor stage, histology, and molecular subtyping—several professional societies have incorporated recommendations for biomarker testing into clinical practice guidelines for the diagnosis and management of NSCLC (Table 1) [4, 7, 14].

Table 1.

Summary of biomarker testing guideline recommendations

Now that guidelines for the treatment of metastatic NSCLC include recommendations for biomarker testing, the application of these recommendations to clinical practice poses some challenges. A key barrier to implementing biomarker testing in metastatic NSCLC is ensuring adequate tissue acquisition and processing for molecular analyses. This acquisition involves a series of decisions that require coordinated efforts and communication across multiple health care disciplines (Fig. 1) [4, 7, 19, 20, 22, 23]. For this reason, a series of three multidisciplinary advisory boards were convened with clinicians from academic institutions (n = 5), community settings (n = 5), and integrated health systems (n = 3) to identify best practices in implementing evidence-based recommendations to guide biomarker-driven therapies. This report summarizes feedback from the three advisory boards and proposes how barriers to implementing biomarker testing might be addressed from the perspectives of advisors in medical oncology (n = 13), nursing (n = 2), molecular and surgical pathology (n = 8), interventional pulmonology (n = 5), interventional radiology (n = 2), and thoracic surgery (n = 3).

Figure 1.

Patient experience with lung cancer. The diagram presents an overview of what a typical patient may experience from the time the patient presents with symptoms of lung cancer through the course of the treatment plan. The three main stages of the patient journey are clinical investigation, diagnosis, and treatment [4, 7, 15–21]. This figure also shows the corresponding health care professionals at each stage. Stages in the NSCLC patient journey emphasize the importance of coordination between different disciplines for the diagnosis and treatment of NSCLC within the framework of a biomarker-driven approach.

Abbreviations: CT, computed tomography; EBUS, endobronchial ultrasound; ER, emergency room; FNA, fine needle aspiration; IR, interventional radiologist; NSCLC, non-small cell lung cancer; PCP, primary care physician; PET, positron emission tomography; TS, thoracic surgeon.

Implementing Guidelines in Practice: Challenges and Recommendations

Molecular profiling has become essential for the treatment of patients with metastatic NSCLC across both academic and community-based practices. The demands of managing a medical practice and the necessary focus on efficiency pose salient challenges for the implementation of molecular testing. For the advisory boards, the topics of guideline interpretation, multidisciplinary communication, and tissue acquisition were presented in broad terms, with specific details on associated challenges provided by the advisors through question-and-answer sessions and open discussion. The challenges identified included (a) staying current with evolving practice standards, (b) lack of communication between relevant stakeholders, and (c) optimizing tissue acquisition and processing. Health care expenditures associated with addressing these challenges are also a concern but are outside the scope of this paper. This document serves as the formalized process for reaching conclusions on how these barriers should be addressed, from the perspective of representatives of academic, community-based, and integrated health care systems. We included the authors’ collective recommendations to assist in both identifying and overcoming these barriers (Table 2).

Table 2.

Challenges identified during advisory boards and summary of recommendations

Challenge: Staying Current With Rapidly Evolving Practice Standards

Diagnosis and treatment options for cancer subtypes, including those specifically for lung cancer, are rapidly expanding at a pace that is becoming increasingly difficult to manage for most clinicians. The list of available biomarker-driven therapies for cancer patients is also growing quickly. It has been estimated that as of 2014, more than 320 biomarker-driven agents are either currently in practice or in clinical trials for lung cancer patients [24]. In addition to established oncogenic mutations in lung cancer, in recent years, several new actionable driver mutations have been characterized in lung adenocarcinoma and squamous cell carcinoma [25]. Furthermore, many drugs targeting the same biological pathway can have several potential predictive biomarkers or assays. Learning this litany of biomarkers and associated therapies and their appropriate clinical uses may add confusion for a community physician tasked with managing multiple tumor types. Those performing diagnostic procedures to obtain adequate tissue for molecular testing need to be cognizant of the issues that might affect tissue acquisition. In addition, pathologists need to select and qualify an appropriate tissue sample that will be tested for analysis and, in many cases, select the laboratory that will receive the specimen.

Recommendation 1A: Practitioners Should Consider Promoting a Local Physician “Champion” to Educate Colleagues in Their Region or Community

The recent discovery and proliferation of relevant genomic information in NSCLC highlights the importance of educating all participating practitioners. Considering that many participating health care providers (primary care physicians, pulmonologists, surgeons, interventional radiologists) often initially evaluate patients and control entry into diagnostic and treatment options, it is essential that these practitioners are aware of the current College of American Pathologists (CAP)/International Association for the Study of Lung Cancer (IASLC)/Association for Molecular Pathology (AMP) guidelines [4]. We recommend promoting a local physician “champion,” someone who is not only motivated to educate health care providers in their region or community but also who is able to act as a recognized lung cancer resource to the health care team. Use of online resources such as the U.S. Food and Drug Administration (FDA) “List of Cleared or Approved Companion Diagnostic Devices” [26], the National Comprehensive Cancer Network (NCCN) [7] guidelines, and My Cancer Genome (Vanderbilt-Ingram Cancer Center, Nashville, TN, http://www.mycancergenome.org) can provide useful data to health care providers regarding advances in NSCLC biomarkers in addition to other tumor types. The person in the physician champion role would undertake careful assessment of the currently available clinical data and enter into strategic dialogue with laboratory personnel, including the clinical pathologist, to ensure that important tissue requirements are recognized. This role may best be served by a medical oncologist already familiar with the relevant scientific data and processes; however, providers from other disciplines could also function in the role. For molecular testing, we suggest identifying proceduralists within the multidisciplinary team who are committed to lung cancer patients and who regularly obtain sufficient tissue during the initial biopsy.

Recommendation 1B: Formal Venues for the Communication of Biomarker Education Should Be Established

Although the optimal format for biomarker education is not clearly established, communication in formal venues such as multidisciplinary tumor boards, cancer conferences, continuing medical education programs, and journal clubs is thought to be effective. Regularly occurring multidisciplinary clinics and tumor boards should be encouraged in the community setting by strengthening institutional incentives and adapting to the realities of physician schedules. Holding meetings before office hours or during lunch (e.g., at 7 am or 12 pm) and providing continuing medical education credit may facilitate tumor board attendance and generate patient referrals when multidisciplinary clinicians come together to discuss patient cases. In addition, informal settings (e.g., phone calls, hospital interactions) are also likely to be useful. Within integrated health networks, standardized implementation of guidelines usually begins with one center; however, that center may lack the leverage to instigate implementation of the guidelines throughout the network. Establishing quality measures and engaging with individual centers within each system may overcome challenges in adherence to an institution’s molecular testing pathway. Education about the importance of molecular testing should extend beyond the cancer care team and into the community of hospitalists and primary care physicians who may refer patients to hospice without recognizing that biomarker-driven therapy might be an option. NCCN guidelines recommend TKIs for patients with EGFR-activating mutations or ALK rearrangements, regardless of performance status [7]. Consequently, molecular testing and biomarker-driven therapies are important to consider for all appropriate patients, even those for whom cytotoxic chemotherapy is not an option due to poor performance status.

Regularly occurring multidisciplinary clinics and tumor boards should be encouraged in the community setting by strengthening institutional incentives and adapting to the realities of physician schedules.

Challenge: Lack of Resources and Communication Between Stakeholders

Executing treatment plans in the era of biomarker-driven therapy requires coordination of care across all disciplines, including medical oncology, nursing, pathology, pulmonology, interventional radiology, and thoracic surgery. Academic medical centers encourage these interactions by virtue of shared spaces, occasional multidisciplinary clinics, and conferences. Although many community-based practices also hold conferences, there may be additional hurdles to maintaining such collaboration. Many physicians (including primary care physicians, hospitalist physicians, pulmonologists, and radiologists) who perform the initial evaluation of patients or biopsies may not foresee the need to consider molecular testing upfront. Laboratory personnel, such as pathologists, may not have the time or resources required to ensure both adequate tissue acquisition and appropriate molecular testing when indicated. The fragmentation that can accompany the patient’s initial assessment, the biopsy procedure, the interpreting physician’s assessment, and the treating clinician involvement is a challenge in the effective communication of complex clinical decision making and how it relates to biomarker testing.

Laboratory challenges have some unique aspects that are worth mentioning. Local laboratories, for instance, may not have the infrastructure to routinely process, store, and prepare slides for molecular testing analysis. If an outside laboratory is used, further confusion may exist regarding which outside laboratory will perform the testing, how this information will be reported back to the treating oncologists, and whether such testing will affect treatment or outcomes. Local pathology groups need to not only account for variability in provider-specific practices but also make decisions about small sample sizes with limited clinical contextual data. In the absence of reflex testing, the treating physician may not feel comfortable ordering the correct molecular test in cases in which the applicability of biomarker-driven therapy is unclear.

Although the advances among patients with treatable driver mutations are quite promising, these patients still represent a minority of cases. As such, these advances may not have fully captured the attention of clinicians and the larger cancer community. The result of these factors is less coordinated care among all relevant stakeholders that can fragment the process anywhere from tissue acquisition to delivery of biomarker-driven therapy.

Recommendation 2A: Every Patient Suspected of Having Advanced-Stage Lung Cancer Should, Ideally, Be Evaluated by a Multidisciplinary Team

We stress the importance of multidisciplinary tumor boards to facilitate coordinated care. By having these cross-functional conversations upfront during the initial clinical investigation stage of the patient’s journey (Fig. 1), the health care team is able to work together to discuss patient-specific diagnostic strategies and potential downstream treatment options. Standardized tumor boards and thoracic oncology clinics help coordinate care and bring awareness of the patient case history to the multidisciplinary health care team. These settings not only afford the opportunity to educate others about current guidelines but also are a chance to operationalize the process of molecular testing, which includes the potential for reflex testing.

Recommendation 2B: Each Institution Should Establish a Molecular Testing Policy That Covers Reflex Testing

Reflex testing is a testing policy that does not require physicians to order the required tests for each individual patient; this type of “standing order” may help expedite the testing process [4]. Unlike the automatic biomarker evaluation of breast cancer specimens, obtaining tissue from the lung can be more complex. Molecular status may not always be ordered or reported in the pathology report for metastatic lung adenocarcinoma. Importantly, CAP/IASLC/AMP molecular testing guidelines do allow for consideration of biomarker testing in early stage patients if agreed by the local institution’s multidisciplinary team [4]. Even when staging information is not explicitly made available to pathologists, there are some instances in which the stage may be inferred by the pathologist to facilitate reflex testing. If a pathologist sees NSCLC with an adenocarcinoma histology from a known liver biopsy, for example, then the stage is metastatic by definition, and molecular testing is recommended by major clinical practice guidelines [4, 8]. Similarly, cytology specimens obtained from malignant pleural effusions are also defined as metastatic according to the TNM staging system (seventh edition) [16]. Should pathologists feel they do not have sufficient information to justify a pathology-initiated molecular test, practice ought to include issuance of an addendum to the pathology report indicating whether there is sufficient tissue or whether a repeat biopsy should be considered for molecular analysis. This addendum should serve to bring molecular testing to the attention of the medical oncologist receiving the pathology report.

Recommendation 2C: Nurse “Navigators” May Help Streamline Patient Care and Facilitate Consistent Communication Among Multidisciplinary Teams

The CAP/IASLC/AMP guidelines indicate that results of molecular testing should be available within 10 working days [4]. Given the time and resource constraints of physician practices, we identify the potential role of a nurse “navigator” within the health care team to help streamline the process of molecular testing if possible. Determining the specific role of the nurse navigator is essential to ensure optimal performance of the entire multidisciplinary team. Responsibilities may include data entry, facilitation of communication between specialties, organization of patient cases for tumor boards, survivorship care, and clinical outcomes assessment. Recent data suggest that integration of a nurse navigator early in the care of cancer patients improves patient experience and reduces problems in care [27].

Recommendation 2D: Electronic Health Records Should Be Maintained and Shared Among the Multidisciplinary Teams

The importance of a shared electronic health record among participating practitioners cannot be understated. Electronic health records not only allow secured sharing of patient information but also can be used to develop interventions and applications that help streamline the process of molecular testing. Electronic or paper forms (e.g., the molecular consultation form used by the U.S. Department of Veterans Affairs) can be routed through the ordering, tissue collection, and pathology groups to help facilitate communication about the purpose of the biopsy and to facilitate getting the entire multidisciplinary team “on the same page.” Check boxes should be used to capture clinical impressions and disease stage, along with a prioritized list of molecular tests to be ordered.

The fragmentation that can accompany the patient’s initial assessment, the biopsy procedure, the interpreting physician’s assessment, and the treating clinician involvement is a challenge in the effective communication of complex clinical decision-making and how it relates to biomarker testing.

Challenge: Optimizing Tissue Acquisition and Processing

The majority of cases of metastatic NSCLC are diagnosed based on small biopsies that are often not of sufficient quantity or quality for molecular testing [28]. Maximizing the amount and quality of tissue for molecular testing poses several hurdles. First, the person who performs the procedure in the community (e.g., pulmonologist, interventional radiologist) may not be as familiar with recent guidelines supporting molecular testing and thus may not prioritize tissue collection. Alternatively, they may feel that the risks of additional tissue acquisition may pose undue risk to the patient.

In addition, it remains unclear which procedural techniques are superior to others for diagnosis and molecular testing. Although core biopsies have traditionally been thought to be superior to cytological specimens, recent data suggest a 98% success rate for EGFR and KRAS testing on pure cytological specimens with the capability to make a tissue block from fine needle aspirations (FNAs) and thoracentesis [29]. Consequently, decisions related to initial tissue acquisition often involve weighing the risks of performing an invasive procedure more likely to yield adequate tissue versus a less invasive procedure that may be easier and pose less risk for the patient.

Future laboratory advances will hopefully allow lower quantities of specimens to suffice, whereas evolving tissue acquisition techniques such as endobronchial ultrasound and electromagnetic navigational bronchoscopy may minimize biopsy risks to patients while enhancing diagnostic yield. Many pulmonologists, however, may not be adept or familiar with these procedures. The lack of prospective data that address optimal techniques for computed tomography-guided or bronchoscopic tissue acquisition amplifies the difficulties. It remains unclear, for example, how many FNA passes of a lymph node are required to yield the appropriate malignant cells to obtain adequate molecular information.

Finally, lack of communication between proceduralists and pathologists can yield inadequate tissue acquisition that may necessitate repeated biopsies and prolong time from diagnosis to biomarker-driven therapy.

Recommendation 3A: Tissue Acquirers and Pathologists Should Communicate Effectively to Ensure That Tissue Obtained for Molecular Testing Is of Sufficient Quantity and Quality

Onsite assessment of tissue quantity and quality by pathology/cytology should be implemented whenever possible to ensure optimal communication between tissue acquirers and pathologists. Rapid onsite assessment can substantially reduce the risk of obtaining inadequate tissue samples and help align the goals of tissue acquisition that will expedite molecular testing. When establishing a diagnosis of NSCLC without rapid onsite evaluation, current guidelines suggest obtaining at least three transbronchial needle aspiration samples [14].

Recommendation 3B: Decisions on the Optimal Diagnostic Procedure for Molecular Testing Should Be Individualized and Include Risk-Benefit Analysis

This recommendation reinforces the importance of a physician champion (described in recommendation 1A) to educate others about both the clinical relevance of molecular testing and the best ways to optimize tissue acquisition safely. Awareness among tissue acquirers of downstream testing will help ensure that adequate tissue is obtained and will help facilitate appropriate molecular testing by the pathology team. Although specific criteria for sample size and tumor cell content have been published [30], questions regarding the optimal approach to yield this information are not clearly defined. These issues should be addressed in a multidisciplinary fashion on a continual basis. It may be recognized that the first biopsy is often the most important; however, there is no consensus about the optimal diagnostic procedure for molecular testing. Procedural decisions should be individualized, with the risks of the procedure weighed against the benefits of maximizing tissue and quality of analytes (e.g., DNA).

Recommendation 3C: Efforts Should Be Made to Ensure Timely Identification of Actionable Biomarkers

The CAP/IASLC/AMP molecular testing guidelines recommend prioritizing EGFR and ALK testing for lung cancer patients with metastatic adenocarcinoma [4]. NCCN clinical practice guidelines recommend multiplex and next-generation sequencing (NGS) for EGFR mutations and/or ALK rearrangements for lung cancer patients with metastatic non-squamous cell carcinoma [7]. Sequentially testing individual biomarkers has also been suggested as a way to enhance the efficiency of molecular testing and to preserve tissue samples; however, testing for biomarkers one at a time may result in longer and potentially unacceptable turnaround times [31]. Although NGS offers the potential to detect a multitude of biomarkers in one assay, these large panels also take time to complete, and consideration must be given to the recommended maximum turnaround time of 10 working days [4]. Molecular testing with NGS or via individual companion diagnostics should not be considered mutually exclusive. NGS panels are useful to screen patients for multiple biomarkers to determine eligibility for clinical trials, and although there may not be clear consensus, some may recognize the use of validated screening methodologies for EGFR mutations and ALK rearrangements while waiting for the results of NGS. This can help ensure that actionable biomarker results from an FDA-approved test are available in time to inform initial treatment decisions; however, reimbursement and tissue availability may limit this particular approach.

Recommendation 3D: Efficient Use of Pleural Fluid May Facilitate Molecular Testing

With regard to pleural fluid, if a thoracentesis was performed first and deemed highly cellular by the pathologist, this should be the initial tissue sent for molecular testing. Sufficient pleural fluid should be provided for molecular testing.

Conclusion

Applying molecular testing guidelines to the clinical care of patients with NSCLC is associated with several key challenges, including education, communication, and tissue handling. In addition, given the long history of dismal outcomes and poor responses to treatment associated with NSCLC, there may be residual pessimism on the part of the medical community. Recognizing these challenges, this report offers the authors’ collective recommendations for overcoming these barriers with the twofold directive of ensuring (a) that all appropriate patients with NSCLC are tested for clinically relevant biomarkers and (b) that molecular testing results are available to prescribers in time for initial treatment decisions. Physician champions can use both formal tumor boards and informal hospital interactions to educate their health care teams about the clinical relevance of testing and the best ways to obtain tissue. During these interactions, tissue acquisition procedures should be individualized in a multidisciplinary fashion with the risks of the procedure weighed against the benefits of maximizing tissue and quality of analytes for potential downstream biomarker testing. Nurse navigators can also help streamline patient care and facilitate consistent communication among all relevant stakeholders. Formalized institutional molecular testing policies, such as reflex testing protocols, and shared electronic health records may facilitate expedited and consistent routing of specimens. As new biomarkers are identified and new diagnostics and treatments become available, the ways in which we as clinicians implement these innovations will need to be continually evaluated to ensure that our patients can realize the full benefits of these advances in modern medicine.

Author Contributions

Conception/Design: Benjamin P. Levy, Marc D. Chioda, Dana Herndon, John W. Longshore, Mohamed Mohamed, Sai-Hong Ignatius Ou, Craig Reynolds, Jaspal Singh, Ignacio I. Wistuba, Paul A. Bunn Jr., Fred R. Hirsch

Provision of study material or patients: Dana Herndon

Manuscript writing: Benjamin P. Levy, Marc D. Chioda, John W. Longshore, Mohamed Mohamed, Sai-Hong Ignatius Ou, Craig Reynolds, Jaspal Singh, Ignacio I. Wistuba, Paul A. Bunn Jr., Fred R. Hirsch

Final approval of manuscript: Benjamin P. Levy, Marc D. Chioda, Dana Herndon, John W. Longshore, Mohamed Mohamed, Sai-Hong Ignatius Ou, Craig Reynolds, Jaspal Singh, Ignacio I. Wistuba, Paul A. Bunn Jr., Fred R. Hirsch

Disclosures

Benjamin P. Levy: Celgene, Genentech, Pfizer, Biodesix, AstraZeneca, Boehringer Ingelheim (C/A), Genentech, Eli Lilly (H); Marc D. Chioda: Pfizer (E, OI); Dana Herndon: Pfizer (C/A); John W. Longshore: Bristol-Myers Squibb, Genentech, Pfizer, Roche (C/A), Roche (RF); Mohamed Mohamed: Pfizer (C/A); Sai-Hong Ignatius Ou: Pfizer, Roche (C/A), Roche, Pfizer, AstraZeneca, Clovis, Ignyta (RF), Roche, Boehringer Ingelheim (H); Craig Reynolds: Lilly, Genentech, Boehringer Ingelheim, Celgene, Bristol-Myers Squibb (C/A, H); Ignacio I. Wistuba: Bristol-Myers Squibb, GlaxoSmithKline, Boehringer Ingelheim, AstraZeneca, Genentech/Roche (C/A), Oncoplex, HTG Molecular, Genentech, Merck, Jounce (RF), Boehringer Ingelheim (H); Paul A. Bunn Jr.: Bristol-Myers Squibb, Amgen, Daiichi, Celgene, AstraZeneca, Genentech, Lilly, Novartis (C/A); Fred R. Hirsch: Lilly/Imclone, Bristol-Myers Squibb, Pfizer, AstraZeneca, Genentech (C/A), Lilly/Imclone, Celgene, Genentech, Amgen, Clovis (RF). The other author indicated no financial relationships.

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board