Associations of ctDNA Clearance (CL) During Neoadjuvant Tx with Pathological Response and Event-free Survival (EFS) in pts with Resectable NSCLC (R-NSCLC): Expanded Analyses from AEGEAN

Background

• In the phase 3 AEGEAN trial, perioperative durvalumab (D) + neoadj CT significantly improved the primary endpoints of EFS and pathological complete response (pCR) vs neoadj CT alone in pts with R-NSCLC.

Aim

To report exploratory analyses for associations of ctDNA CL during neoadj Tx with pCR, major pathological response (MPR) and, for the first time, EFS using data from an expanded cohort of all biomarker-evaluable pts at the second interim EFS analysis.

Methods

• AEGEAN is a double-blind PBO-controlled study (NCT03800134). 

• Adults with Tx-naïve R-NSCLC (stage II–IIIB[N2]) and ECOG PS 0/1 were randomised (1:1) to receive neoadj platinum-based CT + D or PBO IV (Q3W, 4 cycles) before surgery (Sx) followed by D or PBO IV (Q4W, 12 cycles) after Sx. 

• Efficacy was assessed in the mITT population, which excluded pts with known EGFR/ALK mutations. 

• EFS was evaluated by BICR (RECIST v1.1) and pCR/MPR were evaluated centrally (IASLC). 

• ctDNA analysis was performed on plasma collected before each neoadj Tx cycle and before Sx using pt-specific tumour-informed assays.

Results

• ctDNA was evaluated in 1268 samples from 283 mITT pts (D arm, n=142; PBO arm, n=141). 

• Among ctDNA+ pts at baseline (89.6%), all pts who achieved pCR and >93% who achieved MPR had ctDNA CL at C4D1. 

• In both arms, lack of early ctDNA CL identified pts who had a low probability of achieving pCR (negative predictive value [NPV] ≥89% at C2D1). 

• At pre-Sx, pts with vs without ctDNA CL had better EFS outcomes (D arm: HR, 0.26; 95% CI, 0.13–0.54; PBO arm: HR, 0.47; 95% CI, 0.26–0.84)

• And pts in the D vs PBO arm had trends for longer EFS regardless of ctDNA CL (ctDNA CL[+]: HR, 0.50; 95% CI, 0.24–1.05; ctDNA CL[-]: HR, 0.84; 95% CI, 0.47–1.48). 

• Furthermore, pts with ctDNA CL had longer EFS regardless of pCR status compared to pts without either ctDNA CL or pCR who had the worst EFS outcomes (D arm: ctDNA CL[+], pCR[+]; HR, 0.14; 95% CI, 0.04–0.48; ctDNA CL[+], pCR[-]; HR, 0.35; 95% CI, 0.16–0.76; PBO arm.

• ctDNA CL identified more pts with improved EFS than pCR status.

Conclusions

Absence of early ctDNA CL may identify pts unlikely to achieve pCR. Pre-Sx ctDNA CL (alongside pCR) warrants further study as a surrogate endpoint of EFS.

Reference

Annals of Oncology (2024) 35 (suppl_2): 1-72. 10.1016/annonc/annonc1623

Perioperative Nivolumab (NIVO) v Placebo (PBO) in Patients (pts) with Resectable NSCLC: Clinical Update from the Phase III CheckMate 77T Study.

Background

• In CheckMate 77T, NIVO showed clinically meaningful improvements in EFS and pCR v PBO in pts with resectable NSCLC. 

Aim

• To report updated clinical outcomes from the study, exploratory outcomes by pCR status, and ctDNA analyses.

Methods

• Pts with resectable stage IIA–IIIB NSCLC were randomized 1:1 to neoadjuvant (neo) NIVO + chemo Q3W (4 cyc) followed by adjuvant (adj) NIVO Q4W (13 cyc) or neo PBO + chemo Q3W (4 cyc) followed by adj PBO Q4W (13 cyc). 

• Primary endpoint: EFS per BICR. 

• Exploratory analyses: efficacy/safety by pCR status, ctDNA clearance (CL; detectable ctDNA at neo tx start to no detectable ctDNA at neo tx end) and recurrence (no detectable ctDNA at adj tx start to detectable ctDNA at last available adj phase assessment).

Results

• Data cutoff - 26 Apr 2024

• Median f/u - 33.3 mo

• NIVO (n = 229) continued to provide EFS benefit v PBO (n = 232) in all randomized pts (2y EFS rates 65% v 44%; HR [95% CI] 0.59 [0.45–0.79]). 

• BL characteristics, including TN stage, were similar between pts with pCR (NIVO 58; PBO 11) or w/o pCR (98; 148) and between tx arms, except a higher percent of pts with pCR had tumor PD-L1 ≥ 1% (NIVO). 

• Landmark EFS from surgery continued to favor NIVO v PBO in pts with pCR (HR [95% CI] 0.59 [0.12–2.91]) or w/o (0.75 [0.51–1.09]). 

• In ctDNA-evaluable pts (NIVO 76; PBO 64), ctDNA CL rates were higher at the end of neo tx in the NIVO v PBO arm (66% v 38%)

• Pts with ctDNA CL had higher pCR rates (NIVO 50% v PBO 12%) than pts w/o (0% v 2%). 

• In pts with undetectable ctDNA at adj tx start (48; 44), ctDNA recurrence rates were lower in the NIVO v PBO arm (8% v 20%); 

• Among pts with pCR (NIVO 26; PBO 5), ctDNA recurrence rates were 4% v 20%, and in pts w/o pCR (22; 39), the rates were 14% v 21%. 

• Grade 3–4 TRAEs were consistent with the previous report (32% v 25%).

Conclusions

In this update, pts treated with NIVO continued to derive clinical benefit v PBO, regardless of pCR status. Exploratory analyses showed greater ctDNA CL in the NIVO v PBO arm, which was associated with pCR benefit. Inversely, ctDNA recurrence, a marker of disease progression, was lower during the adj phase in the NIVO v PBO arm, in pts with or w/o pCR. These data, including comprehensive ctDNA analyses, further support perioperative NIVO as an efficacious tx option in pts with resectable NSCLC.

Reference

Annals of Oncology (2024) 35 (suppl_2): 1-72. 10.1016/annonc/annonc1623.

Adagrasib (ADA) vs Docetaxel (DOCE) in Patients (pts) with KRASG12C-mutated Advanced NSCLC and Baseline Brain Metastases (BM): Results from KRYSTAL-12.

Background

• Pts with KRAS G12C-mutated NSCLC and BM have poor prognosis. 

• ADA, a KRASG12C inhibitor, significantly improved progression-free survival (PFS) and objective response rate (ORR) vs DOCE in pts with previously treated KRAS G12C-mutated advanced NSCLC in the phase 3 KRYSTAL-12 trial (NCT04685135). 

• In CNS evaluable pts, intracranial ORR was improved in pts with baseline BM receiving ADA (40%) vs DOCE (11%). 

Aim

• To report efficacy and safety outcomes in pts with and without baseline BM from KRYSTAL-12.

Methods

• Pts with KRAS G12C-mutated locally advanced/metastatic NSCLC who had previously received both platinum-based chemotherapy and anti-PD(L)1 therapy were randomized 2:1 to receive

• ADA (600 mg BID orally) or 

• DOCE (75 mg/m2 Q3W IV). 

• Pts with treated, neurologically stable baseline BM were eligible. 

• BM were identified using an independent CNS imaging charter and neuroradiologist review. 

• Exploratory analyses included systemic efficacy (ORR, PFS, duration of response, assessed per RECIST v1.1) and safety by baseline BM, and intracranial efficacy (assesed per CNS-adapted RECIST v1.1) in pts with baseline BM, all by blinded independent central review.

Results

• N= 453 pts

• 114 (25.2%) had baseline BM (ADA n = 78; DOCE n = 36)

• Baseline characteristics were similar across treatment and BM groups. 

• Systemic efficacy outcomes (median follow-up 7.2 months) were improved with ADA vs DOCE, regardless of BM status. 

• Treatment-related adverse events (TRAEs) were reported in 93.5% and 85.7% of pts with BM receiving ADA and DOCE, respectively. 

• TRAEs led to discontinuation of ADA and DOCE in 6.5% and 17.1% of pts with BM, respectively.

Efficacy by baseline BM in KRYSTAL-12

	Patients with baseline BM		Patients without baseline BM
	ADA (n = 78)	DOCE (n = 36)	ADA (n = 223)	DOCE (n = 116)
Systemic PFS, median (95% CI), months HR vs DOCE (95% CI)	4.4 (3.1–5.8) 0.7 (0.4–1.2)	2.9 (2.0–6.2) –	5.9 (4.8–7.2) 0.5 (0.4–0.7)	3.9 (2.4–5.6) –
Systemic ORR, n (%) 95% CI	21 (26.9) 17.5–38.2	1 (2.8) 0.1–14.5	75 (33.6) 27.5–40.2	13 (11.2) 6.1–18.4
Systemic DOR, median (95% CI), months	7.4 (4.2–9.7)	5.4 (NE–NE)	8.3 (5.8–18.2)	5.4 (2.8–8.5)

DOR, duration of response; NE, not estimable.

Conclusion

In pts with previously treated KRAS G12C-mutated advanced NSCLC, ADA demonstrated improved efficacy outcomes vs DOCE, including in pts with baseline BM, consistent with benefits observed in all randomized pts in KRYSTAL-12.

Reference

Annals of Oncology (2024) 35 (suppl_2): 1-72. 10.1016/annonc/annonc1623.

Precision Immuno-oncology for Advanced Non-small Cell Lung Cancer (NSCLC) Patients with PD-(L)1 Inhibitors Resistance (PIONeeR): A Phase Ib/IIa Clinical Trial Targeting Identified Resistance Pathways.

Background

• Anti-PD-(L)1 are approved for the treatment of advanced NSCLC. 

• However, most patients (pts) progress during treatment with limited options after anti-PD-(L)1. 

Aim

• The aim of PIONeeR trial was to overcome resistance to anti-PD(L)-1 by targeting 4 major pathways.

Methods

• This open-label, multicenter, controlled randomized study used a Bayesian adaptive design. 

• Advanced NSCLC pts with disease progression after sequential or concomitant PD-(L)1 and platinum-based chemotherapy were eligible. 

• Pts were randomly allocated to 

• Arm A: Durvalumab (Du) + Monalizumab, 

• Arm B: Du + Oleclumab, 

• Arm C: Du + Ceralasertib, 

• Arm E: Du + Savolitinib or 

• Arm D (control): Docetaxel. 

• Primary endpoint - 12-week Disease Control Rate (12-w DCR) by RECIST1.1. 

• Main secondary endpoints - Duration of Response (DoR), Overall Response Rate (ORR), Progression-Free Survival (PFS), Overall Survival (OS) and safety. 

• Interim analyses used a futility boundary of 30%. 

• A combination arm was considered effective if there was a high probability (P ≥ 90%) for 12-w DCR to be ≥ 12-w DCR of arm D.

Results

• N= 114 pts were randomized in arms A, n=28; B, n=3; C, n=32; D, n=31 & E, n=20. 

• Arms B and E were prematurely closed due to lack of efficacy.

• 11 pts enrolled in arm D withdrew prior to treatment. 

• The 12-w DCR was 54.5% (95%CI [34.0%; 74.3%]) in the control arm vs 24.1% [10.7%; 41.0%] in arm A, 0% in arm B, 50% [33.1%; 66.9%] in arm C and 13.6% [3.0%; 30.4%] in arm E; P was 1.2%, and 36.8%, in arm A and C respectively. 

• No significant PFS or OS benefit was seen in experimental arms compared to arm D. 

• However, 5 pts achieved PR by RECIST in arm C with a median DoR of 5.6 months [2.1; 9.7]. 

• No unexpected safety signal was seen. 

Conclusions

With its innovative and adaptive design, the PIONeeR trial was able to explore several options to overcome resistance to ICIs. Although no experimental arm performed better than outcomes observed with docetaxel, some pts had long DoR, suggesting durvalumab combinations can be highly effective. Biomarker work is ongoing to identify patients most likely to benefit from combination treatment.

Reference

Annals of Oncology (2024) 35 (suppl_2): 1-72. 10.1016/annonc/annonc1623.

The Potential of VEGF-A in the Peripheral Blood Serum as a Biomarker of Response in the Addition of Bevacizumab with Chemotherapy-combined Immunotherapy for Metastatic Nonsquamous Non-small Cell Lung Cancer

Background

• Anti-vascular endothelial growth factor (VEGF) agents like bevacizumab have improved of outcome in patients with various cancers in the era of immunotherapy, but predictive biomarkers that select patients who will benefit from the addition of such anti-VEGF agents to immune checkpoint blockade have not been elucidated.

Aim

Levels of VEGF-A (total VEGF-A) and its two variant isoforms VEGF165 and VEGF121 in the peripheral blood serum at baseline before the initiation of treatment were investigated. 

Methods

• A preplanned, prospective biomarker study affiliated with the APPLE study (Trial number: UMIN000037326), which was an open-label, randomized phase 3 trial evaluating the efficacy of the bevacizumab in combination with a Programmed-cell death (PD)-1 pathway inhibitor, atezolizumab, plus platinum chemotherapy in metastatic, nonsquamous non-small cell lung cancer (NSCLC) patients was conducted. 

• N= 152 patients 

• By using newly developed Enzyme-linked immunosorbent assay system, levels of VEGF-A (total VEGF-A) and its two variant isoforms VEGF165 and VEGF121 in the peripheral blood serum at baseline before the initiation of treatment were investigated. 

• Each VEGF values were defined as high if higher than the median and low if lower than the median in all analyzed populations, and the correlations of those levels with treatment response were analyzed.

Results

• It was revealed that in EGFR-WT patients with the low level of VEGFs, especially total VEGF-A, the addition of bevacizumab to atezolizumab and chemotherapy significantly improved the PFS in comparison to those in patients treated with chemotherapy plus atezolizumab (HR, 0.46; 95% CI, 0.26 to 0.82) compared with those with high VEGF (HR, 1.40; 95% CI, 0.78 to 2.51).

Conclusions

The result demonstrates for the first time that measuring VEGF-A and its isoforms in the serum may identify nonsquamous NSCLC patients with EGFR-WT who will obtain improved prognosis by the addition of bevacizumab to the PD-1 pathway inhibitor’s treatment, and further research into a role for VEGF-A in tumor immunity is warranted.

Reference

Annals of Oncology (2024) 35 (suppl_2): S238-S308. 10.1016/annonc/annonc1576

Circulating tTumor DNA (ctDNA) Dynamics and Treatment Responses in Chemotherapy-ineligible Patients (pts) with Unresectable Stage III NSCLC from the Phase II DUART Trial

Background

In the phase 2, open-label, single-arm DUART trial, radiotherapy (RT) followed by durvalumab (D) showed promising results in chemotherapy-ineligible pts with unresectable stage III NSCLC. 

Aim

To report exploratory ctDNA analyses.

Methods

• Pts without disease progression after standard (54–66 Gy) or palliative RT (40–<54 Gy) received D for up to 12 months. 

• ctDNA samples collected at protocol-specified timepoints (prior to D administration at cycle 1 [C1], C2, and C7) were analyzed with a targeted methylation-based test (GRAIL)

• ctDNA detection, dynamics, and associations with progression-free survival (PFS) were assessed. 

• Survival analyses used a Cox proportional hazards model, with p-values calculated via log-rank test.

Results

• N= 216 samples from 93 pts (standard RT: n=47; palliative RT: n=46) from the intent-to-treat (ITT) population. 

• Baseline characteristics of pts in whom ctDNA was evaluated and the ITT population were similar. 

• The proportion of evaluable pts with detectable ctDNA at C1 was 35% (32/91) and was numerically higher after palliative (20/45 [44%]) vs standard RT (12/46 [26%]). 

• Detectable ctDNA at C1 was associated with a trend toward shorter PFS irrespective of RT dose (HR: 1.60; 95% CI: 0.89, 2.86; p=0.11); this trend in PFS was primarily due to the results from pts receiving standard RT (HR: 2.84; 95% CI: 1.16, 6.99; p=0.017). 

• Among pts evaluable for ctDNA clearance (CL), the ctDNA CL rate in pts receiving palliative vs standard RT, respectively, was similar at C2 (4/21 [19%] vs 2/13 [15%]) and higher at C7 (6/14 [43%] vs 0/10 [0%]). 

• Among pts with available samples at C7, estimated median PFS was 25.5 months for pts with undetectable ctDNA (n=26) vs 9.2 months for pts with detectable ctDNA (n=18), irrespective of RT dose (HR:4.43; 95% CI: 1.55, 12.72; p=0.0026).

Conclusions

The presence of ctDNA after RT alone does not clearly associate with prognosis among pts receiving consolidation D, but absence of detectable ctDNA later during consolidation is associated with prolonged PFS. RT followed by D can lead to complete molecular responses and improved PFS. ctDNA monitoring could help identify pts who may benefit from more intensive treatment in future trials.

Reference

Annals of Oncology (2024) 35 (suppl_2): 1-72. 10.1016/annonc/annonc1623