Naming and Staging Lung Cancers

In any clinical trial or study of cancer, researchers must compare how tumors grow and respond to therapy, which requires everyone calling the same thing by the same name. That truism applies to both the pathological subtype of the tumor, small-cell or non–small-cell lung cancer, for example, and the clinical stage, from early to metastatic disease. Both the pathologic classification and clinical staging influence treatment decisions, albeit in different ways. Pathologic classification tells what subtype the cancer is, which often indicates which therapy is most appropriate. Staging tells where the cancer is in the body and whether it has spread beyond the primary tumor to lymph nodes or metastases, which helps determine whether to treat a patient with local therapy (surgery and/or radiation), systemic therapy, or both.

The fast pace of clinical research in lung cancer has complicated both efforts, particularly for the most common category, non–small-cell carcinomas. For this large, heterogeneous group of adenocarcinomas, identifying genetic abnormalities in specific subtypes has led to targeted therapies that improved survival of patients with even advanced disease.

The two most widely used systems are the World Health Organization’s (WHO) pathologic classification of lung tumors and the International Association for the Study of Lung Cancer’s (IASLC) tumor–node–metastasis (TNM) staging system. Those systems are trying to keep abreast of recent research findings. In September 2015, WHO published its first update¹ since 2004, and IASLC announced recommendations for revisions to its 2011 edition, with its eighth TNM edition² slated for publication in 2017.

Progress in Pathological Classification

“WHO took the lead many decades ago in classifying lung cancer pathologies,” said Peter Illei, M.D., a lung cancer pathologist at the Johns Hopkins Kimmel Cancer Center in Baltimore. It published the first issue pathologic classification in 1967, with updates every decade or so. The previous 2004 edition began including some genetic and immunohistochemistry findings, but the next decade of research revealed many more clinically significant molecular distinctions among lung cancer subtypes. In 2011, IASLC, the premier organization for treating and diagnosing lung cancer, and other collaborators revised the system for pathologic classification of lung adenocarcinomas after reviewing evidence-based medicine and genetic studies. “That 2011 revision fundamentally changed the role of pathology classification from relying on histology to also analyzing molecular biology and immunohistochemistry,” said William Travis, MD, a pathologist at the Memorial Sloan–Kettering Cancer Center in New York and editor of the 1999, 2004, and 2015 editions of the WHO classification.

The 2015 WHO classification now includes more genetic testing for driver mutations, such as EGFR and ALK, that are targeted by U.S. Food and Drug Administration–approved therapies. “WHO also did a good job by including additional genetic targets that have targeted therapy in clinical trials,” Illei said. “A promising area that is not addressed is immunotherapy because the data are still coming in.”

Frank Detterbeck, M.D.

One major change is the reclassification of tumors previously classified as large-cell carcinoma, “a term that was historically a wastebasket for poorly differentiated tumors,” Travis said. With the availability of easily applied antibodies for immunohistochemical markers of adenocarcinomas (such as TTF-1) and squamous cell carcinomas (such as p40 or P63), most of these tumors now have more definitive designations, so clinicians have a better idea how to treat the patient.

Another change addresses the issue that most lung cancer patients are elderly and first present with advanced disease that cannot be surgically removed. Those cancers are diagnosed in small biopsies or cytology specimens that may not represent the entire tumor. In recognition of this problem, the classification includes a separate terminology for these small tissue samples and offers guidelines for using special stains and for triaging for molecular analysis. “Fortunately, it works,” Illei said. Clinicians got better at obtaining enough tumor tissue by using image-guided biopsy techniques, he added, and pathologists got better at handling tissue and preserving enough tissue for molecular testing. “In most cases, we can get enough information to put the tumor in the right histologic category and provide the oncologist with genetic information they need for optimizing treatment for the individual patient.”

TNM Staging

Clinical staging has come a long way since the 1940s, when a surgical oncologist developed descriptors for the TNM system. In 1968, researchers arrived at the first international TNM classification system by consensus, without empirical data.³ In 1996 IASLC’s fifth TNM revision was based on a few cases, predominantly surgical, from one institution, accumulated over 20 years. IASLC then established a Staging and Prognostic Factors Committee to broaden the scope of research. For the seventh edition,⁴ that committee examined almost 100,000 cases between 1990 and 2000, involving all modalities of care, in 46 cancer centers in 20 countries: an unprecedented (if retrospective) dataset. The upcoming eighth edition analyzed another 100,000 cases from 1999 to 2010.

The goal of TNM staging is to determine the clinical stage of a patient on the basis of tumor size in centimeters and the extent of spread to lymph nodes or other sites, said Frank Detterbeck, M.D., a thoracic surgeon at Yale School of Medicine in New Haven, Conn. Each T, N, and M designation is graded by size, number, or severity, and the combined parameters determine the clinical stage. Stage 1 or 2 cancers are considered curable with surgery and/or radiation. Metastatic disease, regardless of tumor size or nodes, is stage 4: incurable but still treatable with systemic therapy to prolong survival.

Stage 3 cancers represent a middle ground, where any combination of treatment might be considered. “This is a heterogeneous group of patients,” said Patrick Forde, M.D., an oncologist at Johns Hopkins Kimmel Cancer Center. “Some have a very small primary tumor but lots of nodes; others have larger primary tumors but no nodes. Some have better prognosis than others.”

Recent TNM revisions reflect attempts to better group stage 3 patients according to prognosis based on the retrospective data, moving descriptors between T and M categories, which can also change the clinical stage. In the seventh edition, cases with maligned pleural or pericardial effusions or nodules were moved from T to M category, making the cancer stage 4.⁵ For the eighth edition, diaphragm invasion, which has a poor prognosis, will be upgraded from T3 to T4.

“TNM classifications are important in giving patients a prognosis, a sense of whether they will likely survive for 1 year or 5 years,” said Sylvia Lee, M.D., a medical oncologist at Seattle Cancer Care Alliance in Washington. “But there’s a limitation because the survival curves are based on data that are 5 or 10 years old. Patients with advanced metastatic disease who once would have died within a few months can live for years on targeted therapies. The TNM staging still helps with treatment decisions: Does this patient need surgery or not, radiation or not, systemic therapy or not?”

Staging of Multiple Tumors

Ambiguities in TNM staging can arise when a patient has a second tumor in the lung. Is one then one tumor the primary tumor and the other a metastasis? Or are both primary tumors? “Biopsies cannot reliably answer those questions,” Detterbeck said. In 2014, he surveyed experts and clinicians who regularly evaluate lung cancer patients and found no consistency in how experts classify hypothetical cases of multiple tumors in the lung.⁶ “How we think about these cases affects our approach to treatment and also patient outcome,” he said. “If we think of the second tumor as another primary stage 2 lung cancer, we treat both with curative intent. If we think the second one is a metastasis, we think we can’t cure it and will not use surgery or radiation.” In such cases, he said, clinicians should use imaging to follow the pattern of tumor growth and consult with a tumor board of experts in pathology, radiology, surgery, and medical oncology to reach a collective judgment.

Resolving such ambiguities is important for prospective research studies, Detterbeck said. “If different clinicians classify the same cancer subtype differently, or two different subtypes as the same type, then any data about how these tumors behave or respond to therapies will not be reliable.”

A similar issue affects whether patients have a single or multiple metastases beyond the lungs. In some cancers, a solitary distal metastasis means poor prognosis, but in lung cancer one metastasis is sometimes cured with local surgery. The eighth TNM edition will now distinguish single (M1b) from multiple distal metastasis (M1c, a new category). That distinction is important in clinical trials for both enrolling patients and interpreting results, Forde said. “Most trials report the TNM stage of the patient. If lots of patients in a study of new drug have single metastasis, it could bias results in favor of the therapy. If most have multiple metastasis, it could negatively impact the outcome of the trial.”

Getting Word Out

Most patients, however, do not participate in clinical studies. “They are treated in the community and not by specialists trained in thoracic oncology,” Illei said. By using the updated WHO classification and TNM staging and following treatment guidelines, patients can benefit from recent progress in the field—although, he noted, some lag and disconnect persists. Many patients who should have been tested for EGFR and ALK mutations, for example, were not, whereas many academic research centers go beyond the guidelines and already test for other targets under investigation. Still, the guidelines are as updated as practically feasible in this rapidly advancing field—a boon to clinicians and patients. But with the rapidly increasing number of molecular targets for new targeted agents, new immunotherapies, and new survival curves of recently added therapies, the new WHO and upcoming TNM editions will probably see new revisions before the decade is out.