Temporal trends in clinical and pathologic stage discordance in resectable non-small cell lung cancer.

Abstract

e20068 Background: Neoadjuvant chemoimmunotherapy is increasingly utilized in patients with resectable non-small cell lung cancer (NSCLC), with treatment decisions based on clinical (cTNM) rather than pathologic (pTNM) stage. Therefore, the accuracy of cTNM needs to be better defined. We evaluated the correlation between cTNM and pTNM stage in NSCLC using the National Cancer Database (NCDB), which captures approximately 72% of all patients with cancer cases in the United States. Methods: The NCDB was queried for patients with surgically resected NSCLC diagnosed between 2004 and 2020 with complete clinical (c) and pathological (p) staging information on tumor (T) and lymph nodes (N). Patients with clinical stage IV, cN3, no lymph node examined during surgery, and those who received neoadjuvant therapy were excluded. The discrepancies of cTNM and pTNM were evaluated for all patients, by histological subtype, and year of diagnosis. The histology was subdivided into lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LSCC) and others. Upstaging and downstaging were defined as discrepancies between cTNM and pTNM stages. Pearson’s correlation coefficient (r) was used to assess the correlation between year of surgery and changes in upstaging and downstaging. Results: Among the 325,794 eligible patients, most had cT1 (63.7%), cN0 (88.9%). The median number of lymph nodes examined was 9 with an interquartile range from 5 to 15. The median number of regional lymph nodes examined in patients with cN1 increased from 9 in 2004 to 14 in 2020 (r = 0.89, p < 0.0001). Tumor upstaging was seen in 63,241 (19.4%) patients and downstaging in 14,237 (4.4%), whereas N upstaging was seen in 42,406 (13%) patients and downstaging in 9,552 (2.9%). Upstage in cN0 and cN1 was 13.4% and 15.4% respectively, and downstage in cN1 and cN2 was 21% and 36.1% respectively. The percentage of unsuspected pN2 was 5.3% for the entire cohort, 4.7% in patients with cN0 and 15.2% in patients with cN1. Over time, there was a small variation in the cN0 to pN2, ranging from 3.8% to 5.2% (r = 0.5478, p = 0.02). However, cN1 to pN2 upstaging continuously increased from 8.5% to 20% (r = 0.98, p < 0.001). The percentage of patients with N upstage increased from 12.3% in 2004 to 13.8% in 2020 (r = 0.8892, p < 0.0001). The N downstage decreased from 4.1% in 2004 to 1.9% in 2020 (r = -0.95, p < 0.0001). N upstage was seen in 13.6% of LUAD and 12.4% of LSCC, whereas N downstage was seen in 2.2% of LUAD and 4.6% of LSCC. Conclusions: Although there is a good correlation between cTNM and pTNM stages, a discrepancy between cT and pT stages occurs in 23.8%, and a discrepancy between cN and pN stages occurs in 15.9% of cases. The percentage of unsuspected pN2 increase may be related to the higher number of lymph nodes examined. Clinical decisions on the treatment of resectable NSCLC, especially with the use of neoadjuvant therapy, should consider the discordances between cTNM and pTNM stages.