Lung cancer is the most common cause of cancer deaths in men and women. Prognosis depends on the extent of disease at the time of diagnosis. Accurate staging, assessing the extent of local and distant disease, is necessary to determine resectability and overall prognosis. Surgical resection is the only curative hope for patients with lung cancer. One goal of staging is to classify properly operable patients and avoid surgery in those with unresectable disease. The TNM (tumor, node, metastasis) system is commonly used to classify the anatomic extent of disease. The interactive table below defines the currently accepted staging system for non-small cell bronchogenic carcinoma. Current statistics on 5 year survival are also shown and examples of different stages are illustrated. Surgical options exist for patients up to StageIIIA. In general, those with Stage IIIB and IV are considered unresectable.
Accuracy of CT to detect lymph node metastases
A meta-analysis of 42 studies published between 1980 and 1988 found (using a node size greater than 1.0 cm as abnormal) a pooled sensitivity of 83%, specificity of 81% and accuracy of 81%. The 20% false negative rate is largely due to microscopic metastases to normal sized lymph nodes and the 20% false positive rate is due to enlarged nodes from pre- or coexisting inflammatory disease. Larger series (see table below) have demonstrated even worse results.
In general, lymph node metastases increased in patients with:
- Adenocarcinoma
- T> 3.0 cm in diameters
- Central tumors
Caveats
- Prevalence of metastases in normal sized nodes: 13%
- Frequency of nodal metastases: 21-52%
- False negative rate left mediastinal adenopathy : 70%, Agreement poor: kappa 0.28
| Study | Year | n | Sensitivity | Specificity | |
| McCloud | 92 | 143 | 64 | 62 | 1.68 |
| Webb | 91 | 170 | 52 | 69 | 1.67 |
| Grenier | 89 | 85 | 46 | 79 | 2.19 |
| Staples | 88 | 151 | 79 | 65 | 2.20 |
| n | Size | % metastases | Odds |
| 336 | <1 | 13 | 0.15 |
| 57 | 1.0-1.9 | 25 | 0.33 |
| 13 | 2.0-2.9 | 62 | 1.63 |
| 6 | 3.0-3.9 | 67 | 2.00 |
| 2 | >4 | 100 | Infinity |
| Note that the frequency of metastases increases with lymph node size. Even though the numbers are small, not all enlarged nodes contain metastases. | ||||||||||||||||||||||||||
|
||||||||||||||||||||||||||
*Agreement (Kappa) from Guyatt GH. Note the poor sensitivity and agreement for left sided mediastinal lymph nodes.
M1: Evaluation for Metastases
Less than 50% of patients undergoing curative resection for bronchogenic carcinoma survive 5 years. Most deaths are due to local recurrence or disseminated disease. Analysis of the causes of death within a year of surgery suggest that a third of patients have metastases. Likely sites of metastases include bone, brain, liver, and adrenal glands.
The probability of metastases depends on the TNM stage and the histologic type. The higher the stage the more frequent the metastases. For example, there is a 12 fold increase in cerebral metastases in patients with T2N1 tumors compared to T1N0.
The frequency of metastases are less for squamous cell carcinomas than the adenocarcinomas. In resected patients with Stage I cancers, metastases occurred in 15% with squamous cell carcinoma vs 27% with adenocarcinomas.
Hillers et al performed a literature review (1966-91) to determine the proportion of patients with potentially operable non-small cell lung cancer that could be spared thoracotomy by a search for extrathoracic metastases. Although only 17 studies were suitable for analysis the frequency of metastases were (95% confidence interval)
- Head CT 3.3% (2.1 - 4.4%)
- CT adrenal 4.7% (3.0 - 6.4%)
- Bone scan 9.3% (6.7 - 12%)
- Liver Imaging 2.3% (0.9 - 3.3%)
