Staging of lung cancer: the 8th TNM classification

In 2009, the 7th Edition of the TNM classification of malignant tumours was published [1]. The proposals for the revised T, N and M categories will be implemented in the 8th Edition that is expected for late 2016. These proposals have been developed by the Staging and Prognostic Factors Committee of the ­International Association for the Study of Lung Cancer (IASLC) based on a large prospective database [2]. “It contains almost 100,000 cases and has been built up from 1990 onwards,” explained Wilfried Eberhardt, MD, Department of Medical Oncology, University Hospital Essen, University of Duisburg-Essen, Germany.
The IASLC Lung Cancer Staging Project collects not just anatomical information, but also non-anatomical data. These include patient-related elements (e. g., age, sex, race, smoking history, laboratory analyses, lung function test results), tumour-related elements (e. g., maximum standardised uptake value for tumour and lymph nodes, histolo­gical type, vascular invasion, tumour markers) and environment-related elements (method of detection, treatment, residual tumour, geographic area). Dr. Eberhardt detailed the proposals for the T, N and M descriptors and the new stage groupings for ­patients with NSCLC (Table).

Descriptors and T and M categories in the 7th Edition and as proposed for the 8th Edition. The resulting stage groupings proposed for the 8th Edition are highlighted in bold where changes have been implemented (the 7th Edition stage is given in parentheses).[1,2]

T descriptors

With respect to the T descriptors, the proposals underline the significance of tumour volume, which is one of the major prognostic factors. The tumour diameter can also be used as a surrogate marker for tumour volume. “This has always been part of staging, but it is becoming more and more important”, Dr. Eberhardt said. T1 has been subclassified into T1a (≤ 1 cm), T1b (> 1 to ≤ 2 cm) and T1c (> 2 to ≤ 3 cm). Likewise, T2 now consists of T2a (> 3 to ≤ 4 cm) and T2b (> 4 to ≤ 5 cm). Tumours > 5 cm to ≤ 7 cm have been reclassified as T3, and those > 7 cm belong to the T4 category. “We are taking into account the poor prognosis of larger tumours,” Dr. Eberhardt noted. As is typical of solid tumours, life expectancy decreases with increasing lung cancer size.
Another change pertains to the involvement of the main bronchus, which has been classified as T2 regardless of the distance from the carina. Moreover, partial and total atelectasis/pneumonitis are defined as T2, and diaphragm invasion has been reclassified as T4. Mediastinal pleura invasion has been removed as a T descriptor. On the other hand, various features have been maintained in the new edition, such as the definition of visceral pleural invasion as a T2 descriptor, and the subclassification of parietal pericardium, mediastinal pleura and chest-wall invasion, Pancoast tumour, parietal pleural invasion, and additional nodules in the same lobe as T3 descriptors. Also, involvement of the mediastinum, pulmonary artery, aortic wall, vena cava, vertebral body, trachea and carina, and separate nodules in other ipsilateral lobes, are still classified as T4 descriptors.
As Dr. Eberhardt pointed out, these proposed changes successfully serve the purpose of defining different prognostic groups more clearly. “Whereas the survival curves hardly differed between cT3 and cT4 according to the 7th Edition, the new T categories make for nice separation of all of the staging groups.” This observation applies to both clinical and pathological staging.

N descriptors

The database did not provide enough information to warrant implementation of changes with regard to the N descriptors. For both clinical and pathological staging, the 5-year survival estimates vary distinctly according to the established categories of N0 to N3. “Therefore, the committee decided that the current N descriptors adequately predict prognosis and should be maintained in the forthcoming staging system,” Dr. Eberhardt reported.
However, based on the large heterogeneity of stage III lung cancer, it has been recommended that physicians ­record the number of metastatic lymph nodes (or stations) and further classify the N category using the new descriptors of N1a (single), N1b (multiple), N2a1 (single without N1 – skip), N2a2 (single with N1), N2b (multiple), and N3. “This was felt to be necessary to ­define the heterogeneity of stage III ­patients more correctly in future classifications,” Dr. Eberhardt said. These changes will hopefully provide more data that will enable the committee to implement new recommendations in the next revision.

M descriptors

Important changes have been proposed regarding the M descriptors, with the aim being to define oligometastatic disease. The restriction of analyses to long-term survivors in this group of patients naturally promotes selection bias, and the definition of oligometastatic disease itself has been very vague, with the lack of any evidence-based foundation. “Some scientists suggest only one metastatic lesion, others up to five,” Dr. Eberhardt said.
In the current revision of the staging system, several features continue to be grouped as the M1a category (pleural/pericardial effusions, contralateral/bilateral lung nodules, contralateral/bilateral pleural nodules, or a combination of these parameters), whereas single metastatic lesions in a single distant organ have been newly designated to the M1b category. At the same time, multiple lesions in a single organ and in multiple organs have been reclassified as the new M1c category. “This division can serve as a first step towards providing rational definitions for staging of ­oligometastatic NSCLC,” Dr. Eberhardt stated.
The maintenance of the M1a category is based on the lack of differences in survival according to the four para­meters of pleural/pericardial nodules, contralateral/bilateral tumour nodules, pleural/pericardial effusion, and multiple M1a descriptors. Here, the patient numbers remain relatively small, which also applies to the new M1b category. For survival according to single lesions at single sites, only metastases to the adrenal gland stood out in the first analysis, although this difference did not hold true for other subsets. “In the future, more patients with metastatic disease should be included in the database,” Dr. Eberhardt emphasised. Only the M1a and M1b categories are associated with long-term survival according to the 8th Edition, while the M1c category is not.

Similar staging changes in NSCLC and SCLC

Compared to the 7th Edition, the staging of NSCLC that is being proposed for the 8th Edition is more complicated, as it ­involves 11 categories rather than 7. “However, at the end of the day, these curves separate nicely,” Dr. Eberhardt stressed (Figure). Patients with stage IA1 disease experience an average 60-month survival of 92 %; for stage IVB, at the other end of the range, this rate is as low as 0 %, while patients with stage IVA experience long-term survival at 10 %. This difference might have some implications with regard to prospective clinical trials that use multimodal treatments.

Figure: Overall survival according to the 8th Edition IASCL staging proposals [2]

Figure: Overall survival according to the 8th Edition IASCL staging proposals [2]

For small-cell lung cancer (SCLC), as opposed to NSCLC, stage IV is not being divided into subcategories due to insufficient patient numbers. Other than this, the same subsets have been defined as for NSCLC, and these differ clearly with regard to survival. “The prognosis of SCLC patients based on staging is comparable to that in patients with NSCLC.”
As Dr. Eberhardt pointed out, the participation of new centres in the next staging round would be greatly welcomed. “We desperately need more ­patients, and we would appreciate it if additional centres joined us.”

REFERENCES

  1. Sobin LH et al., TNM Classification of ­malignant tumours, 7th edition. November 2009, Wiley-Blackwell, ISBN: 978-1-4443-3241-4
  2. Goldstraw P et al., The IASLC Lung Cancer Staging Project: Proposals for the Revision of the TNM Stage Groupings in the Forthcoming (Eighth) Edition of the TNM Classification for Lung Cancer. J Thorac Oncol 2016; 11(1): 39-51