6 Clinical Data in the Neoadjuvant Setting

6.1 ICI Monotherapy

Neoadjuvant chemotherapy results in a survival benefit com- pared to surgery alone (HR 0.84) [33]. Pathologic response is a surrogate of survival in this setting [34]. In a small trial of 160 patients, major pathologic response (MPR, defined as ≤ 10% viable tumour cells) occurred in 19% of cases. Interestingly, there was no correlation with radiological response [35].

Forde and colleagues performed a window of opportu- nity trial administrating two doses of the PD-1 inhibitor nivolumab to patients with stage I-IIIa NSCLC, planned for curative surgery. Nivolumab 3 mg/kg of bodyweight was administered 2 weeks apart, with surgery planned 4 weeks after the first dose. Primary endpoints were safety and fea- sibility. Twenty-two patients were enrolled and received at least one dose of nivolumab. One patient revealed to have small-cell carcinoma and was withdrawn from the study. There were no new safety signals compared to previous stud- ies. Adverse events were experienced in 23% of patients, only one was grade 3 in terms of pneumonitis after the first cycle of treatment. No delay or omission of surgery occurred, nor any grade 4 or 5 adverse events. Of the 21 patients, 20 achieved complete resection of the tumour. Although only 10% (2 patients) had a radiologically par- tial response between nivolumab and surgery, MPR was observed in 45% (nine patients) including two patients with minor radiological progression. There was a correlation between tumour mutational burden and pathologic response measured in 11 patients with sufficient pretreatment tissue available for sequencing (Spearmans rho − 0.75; p = 0,008). There was no clear correlation with PD-L1 expression as responses occurred in PD-L1 positive and negative tumours as well. Nine patients (three with and six without a major pathological response) could be evaluated for the expan- sion of T-cell clones. In patients with a MPR, there was     a higher frequency of T-cell clones shared between intra- tumoral and peripheral compartments with a higher clonality as a surrogate for an effective immune response [36]. While these results are undoubtedly striking, one has to consider the small number of patients in this trial, making it only hypothesis generating.

In a larger, single-arm trial, patients with stage IB to IIIB NSCLC received two cycles of the PD-L1 inhibitor atezolizumab (1200 mg fixed dose days 1 and 22). Primary endpoint was the percentage of patients achieving an MPR. Secondary endpoints included safety and correlation with PD-L1 expression, TMB, and gene expression signatures. In a recent publication, the results of the first 101 of 180 planned patients were reported. Ninety patients had sur- gery—8 of these were excluded due to the detection of acti- vating alterations in the tumour (EGFR and ALK). Interest- ingly there was no MPR in these patients, which underscores the above-mentioned observation that ICI monotherapy is ineffective in NSCLC with these driver alterations. Of 80 remaining patients, 15 (18%) achieved an MPR, including 4 patients with a complete pathological response. MPR was numerically higher in the PD-L1 positive group (29% and 8%, respectively; p = 0.055). Median TMB was 10.4 muta- tions per megabase and no correlation between TMB and MPR was observed. There were two deaths not related to treatment and 6% Grade 3 or 4 treatment related adverse events [37]. It remains elusive, why MPR was lower with atezolizumab compared to nivolumab. It is in the range achieved with chemotherapy. Whether this is a drug effect or due to chance has to be clarified in larger, randomised trials.

6.2 Combination Strategies

In a study by the Spanish Lung Cancer Group, 46 patients with stage IIIA NSCLC (either N2 or T4N0) received carboplatin [area under the curve (AUC) 6]; paclitaxel (200 mg/m2) and nivolumab (360 mg) every 3 weeks for three cycles followed by surgery in a 6- to 7-week window after the last chemotherapy. Resected patients received 240 mg nivolumab every 2 weeks for 4 months followed by 480 mg every 4 weeks for another 8 months result ing in a year of adjuvant therapy. Primary endpoint was progression-free survival (PFS) at 24 months, secondary endpoints were downstaging rate, complete resection rate, response and 3-year OS, among others. Fifty-one patients were assessed for eligibility, of these, 46 were enrolled and received neoadjuvant therapy. Forty-one patients were resected and 37 received adjuvant treatment (73% of those screened). In a preliminary report, responses have been presented, as shown in Table 4 [38]. The high radiological and histological response rates are impressive. PFS and OS data are, however, pending. Safety data were promising with only one grade 4 event, namely neutropenia.

In a smaller neoadjuvant trial with 14 patients (stage IB–IIIA) received carboplatin, nab-paclitaxel, and atezolizumab for four cycles. Response rate in this group was 57%. Eleven patients could be resected. MPR was 50% in these patients and occurred in PD-L1- positive as well as PD-L1-negative patients [39].

In a small two-arm study, therapy-naïve patients with stage I to IIIA (N2) lung cancer received either nivolumab (3 mg/kg on days 1, 15, and 29) or the same regimen with ipilimumab (1 mg/kg on day 1). Surgery was performed 3–6 weeks after the last dose of checkpoint inhibition. The trial endpoint was an MPR rate ≥ 15%. Forty-four patients were enrolled, 39 of these had surgery. The remaining 5 had progression (1 patient), declined surgery (1), had a lack of resectability (1), or were deemed to have a too high a surgical risk (2). In the combination arm, 6 patients (29%) achieved a complete pathological response (cPR) and one an MPR, resulting in an MPR rate of 33%. In the monotherapy arm the MPR rate was 17% (2 cPR and MPR each). Interestingly in this trial there was a correlation of baseline PD-L1 expression with RECIST-Response and MPR. There were no new safety signals. Altogether, five patients had grade 3 or higher adverse events. One patient in the nivolumab arm died of pneumonitis [40].