187O - A multifactorial score to predict surgical complexity of lung resection following neoadjuvant chemo-immunotherapy

Background

Neoadjuvant chemo-nivolumab for stage II-III NSCLC was approved in UK in March 2013. This study develops a preoperative risk score to predict surgical complexity to support better planning and safer delivery of the procedures.

Methods

This is an observational study based on a prospectively collected data of consecutive patients undergoing surgery following neoadjuvant immunotherapy for locally advanced NSCLC (April 23-Dec 24). The complexity score was quantified by the operator according to 4-grades domains: overall complexity, adhesions severity, mediastinal and vascular dissection. An operation was defined as complex if at least one of the domains scored as severely more complex than a standard lobectomy. Logistic regression was used to test the association of patient (PRF) and tumour related factors (TRF) with procedure complexity. Regression coefficients were used to generate proportionally weighed scores.

Results

65 patients undergoing resection following chemo-nivolumab were analysed. Most frequent procedure was lobectomy (86%). A total of 28 cases (43%) were classified as complex. Median operative time was 228 min in complex cases vs. 180 min, p=0.003. 53 started via minimally invasive surgery and conversion was 48% in complex cases vs. 6.7% in non-complex ones, p=0.001. No PRF was found associated with complexity. Amongst TRF nodal c-stage showed a trend of increased complexity for cN2 (63% complex, vs. 25% in cN1 and 39% in cN0, p=0.07). Absence of nodal response was associated with complexity (62% complex among non-responders vs. 29% among responders, p=0.030). 56% of patients with PD-L1 expression>50% had a complex procedure (vs. 28% of those with PD-L150% (Coeff.1.35, p=0.027, score=1). Patients were grouped into risk classes based on their scores (0 to 5, tab 1). The proportion of complex operations increased with higher scores (from 9.1% in patients with score3, p=0.004).

Table: complex operations proportionally increased with the risk scores (from 9.1% in patients with score3, p=0.004).

Conclusions

The proposed risk score may minimise unplanned intraoperative strategy changes, increasing safety. General applicability requires external validation.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.