Patient survival influences new lung cancer staging system

May 21, 2009

Based on on a database with more than 100,000 submitted cases, modifications to the international system for staging non-small cell lung cancer promise to more closely reflect the connection between disease progression and the patient's prospects for survival.

Based on on a database with more than 100,000 submitted cases, modifications to the international system for staging non-small cell lung cancer promise to more closely reflect the connection between disease progression and the patient's prospects for survival.

The International Association for the Study of Lung Cancer (IASLC) has published its recommendations for change in the upcoming seventh edition of the tumor, node, and metastasis classifications this year. Changes to the 12-year-old sixth edition were drawn from the surgically staged case experience at 46 academic hospitals in 19 countries, said Dr. Johnny Vlahos, an honorary senior lecturer at St. George's, University of London.

The effort invested in the new standards contrasts with planning for the last edition, Vlahos noted. Sixth edition classifications were based on just over 5300 surgically staged cases from a single institution in the 1970s, he said.

Revisions were needed because of changes in histology epidemiology, our understanding of tumor spread and outcome data, and improvements in the accuracy of imaging staging, according to Vlahos.

Patient survival data guided the effort. With the new system, clinicians will be able to say confidently that a patient with a correctly staged T1a lung cancer will have a 77% chance of surviving five years. For stage T2a cancer, survival probability drops to 58%, and patients with metastatic T3 cancer have less than a one in three chance of living five more years.

The new standards generally predict longer survival for patients with a given set of clinical findings than the current classifications.

"We are aware -- conservatively -- that patients previously suspected to have a poor prognosis may have a better prognosis than originally thought," Vlahos said in a presentation at the 2009 International Symposium on Multidetector-Row CT.

Vlahos described the new standards: Tumor (T) classifications will be modified so T1 staging will now consist of T1a for nodules of less than 2 cm and T1b for those from 2 to 3 cm.

The T2 stage will be split into two subcategories: T2a for 3 to 5-cm lesions and T2b for 5 to 7-cm lesions. The putative T2c stage for lesions larger than 7 cm will actually be migrated to a T3 cancer.

The T3 migration was decided because of patient five-year survival rates. These indicated that 77% of patients with T1a stage will survive five years while only 49% T2b cancer patients will survive that long. The 35% rate for T2c category patients is similar to the 31% survival experience for T3 patients, so the two categories were merged.

For metastatic disease, patients with a nodule in the same lobe as the primary tumor tend to survive longer than patients with T4 disease, so they will now be staged with T3 disease. Patients formerly classified with T4 because of malignant pleural disease, have only an 11% chance of surviving five years. They will be classified with M1a cancer. Cases involving the contralateral lung and previously staged as M1 disease have a similar survival profile and will also be staged as M1a. Metastatic disease outside the lung (formerly considered M1 as well) will be called M1b. Cases now staged as M1 cancer and involving cancer that has spread to the nonprimary lobe of the ipsilateral lung will be staged as T4 cancers in the new system.

"In summary, M1a disease is intrathoracic metastatic disease, and M1b is now extrathoracic metastatic disease," Vlahos said.

The N descriptions for nodular involvement have not changed because exploratory analyses performed to simplify nodal staging were not able to be validated across different populations, he said. Prospective trials have been suggested to address that shortcoming.

Limitations of the new classification system include its developers' reliance on retrospective analysis of survival. It does not subclassify disease based on new patterns of management, such as isolated brain or adrenal metastases that are increasingly aggressively managed in select cases. Similarly, they felt the distinction of bulky or small-volume N2 disease does not merit separate consideration. The new system does not address specific tumor types, such as the ground-glass nodules due to bronchoalveolar cell carcinoma and its multifocal variants.

"That probably deserves a separate categorization that considers the spectrum of disease represented among these non-small cell lung cancer classifications," he said.

The reclassification of disease into adjusted T and M descriptions will contribute to substantial stage migration. The general division between former stage IIIa and IIIb designations, which have implications for surgical decision making, will be less definite, he said.

Vlahos and colleagues at hospitals associated with St. George's, University of London studied how the new system may lead to stage migration. Based on 166 consecutive lung cancer patients, they found the proportion of patients classified with stage IIIA or less disease increased from 30.7% to 34.9%. The number of patients staged with IIIB cancer decreased 61.5%, and the number of stage IV cancer cases rose 24%.