AACR: AstraZeneca offers first look at Imfinzi's resectable lung cancer showing, fueling a key debate

AstraZeneca has offered the first look at the benefit of a PD-1/L1 inhibitor when used both before and after surgery in resectable non-small cell lung cancer. But the data raise more questions than answers.

PD-1/L1 inhibitors are already approved for use either before or after surgery in early-stage NSCLC. Industry watchers were looking to AstraZeneca’s phase 3 AEGEAN trial for early clues about whether both treatment settings are necessary to achieve optimal outcomes.

But those who were hoping for a big treatment effect and a conclusive answer may be disappointed.

Using Imfinzi before and after surgery pared down the risk of disease recurrence, progression or death by 32% in patients with resectable stage 2 to 3b NSCLC, according to the AEGEAN data presented at the AACR annual meeting 2023. Imfinzi conferred that event-free survival (EFS) benefit as a pre-surgical addition to neoadjuvant chemotherapy, followed by Imfinzi alone post-surgery in the adjuvant setting. The regimen’s comparator was neoadjuvant chemo alone.

The 32% figure, however, is lower than the 37% EFS risk reduction Bristol Myers Squibb’s Opdivo initially posted in the CheckMate-816 trial. In that study, the PD-1 inhibitor was only used in the neoadjuvant setting alongside chemotherapy for early-stage NSCLC.

In a longer-term follow-up, Opdivo’s magnitude of benefit fell slightly but still came in at 32%.

In CheckMate-816, neoadjuvant Opdivo was given for three cycles. In AEGEAN, Imfinzi was given for four cycles before surgery and then up to 12 cycles after surgery.

Taken at face value, the results from the two trials suggest that Imfinzi’s efficacy in both the neoadjuvant and adjuvant settings compares with Opdivo's in neoadjuvant alone. If this reading proves true, sales prospects for the anti-PD-1/L1 class could take a hit because it would mean patients need shorter treatment periods for positive outcomes.

But in an interview with Fierce Pharma, John Heymach, M.D., Ph.D., chair of MD Anderson Cancer Center’s thoracic/head and neck cancer practice, and lead investigator of the AEGEAN study, cautioned against comparing the two trials directly because of several differences.

The necessity of adjuvant treatment

For one, AEGEAN enrolled patients with stage 2 to 3b cancer while the Opdivo trial included stage 1b to 3a. And to further complicate the comparison, the staging standards between the two studies were also different, Heymach noted.

Although Imfinzi’s EFS win spanned different subgroups, the size of the benefit was smaller in stage 3b patients, reaching a 17% risk reduction, Heymach noted.

Including different cancer stages in a trial is a double-edged sword, Heymach explained. On one hand, the likelihood of recurrence goes up in later-stage tumors, potentially giving post-surgery adjuvant therapy more room to show a risk reduction. But on the other hand, the chance for a complete tumor resection during surgery is lower for more advanced tumors, hurting prospects for preventing recurrence.

In Imfinzi’s case, the lower complete resection rate appeared to have played a bigger role in the trade-off.  

On the other primary endpoint of AEGEAN, Imfinzi did perform better at clearing signs of cancer in resected tissues. Altogether 17.2% of patients achieved the so-called pathological complete response (pCR) after receiving neoadjuvant Imfinzi and chemo, versus 4.3% for chemo alone.

Investigators recorded a pCR improvement for stage 3b patients as well, but the magnitude was smaller at 10.2% for the Imfinzi arm versus 3.1% for control, Heymach noted.

By comparison, in CheckMate-816, Opdivo’s chemo combo delivered a larger pCR rate of 24%, versus 2.2% for neoadjuvant chemo alone.

  pCR Median EFS


PD-(L)1 Control Delta PD-(L)1 Control Risk reduction
17.2% 4.3% 12.9% Not Reached 25.9 mo 32%

Bristol Myers

24% 2.2% 21.8% 31.6 mo 21.8 mo 37%








Besides the patient population difference, Heymach suggested that Imfinzi’s EFS benefit might expand over time because of the neoadjuvant-plus-adjuvant design. The AEGEAN trial hit its goal at an interim analysis after a median follow up of 11.7 months. At that time, only about 65% of patients had actually started their adjuvant treatment.

“We still haven’t seen the benefits of the full adjuvant period because of how quickly the study was read out as positive at the first analysis,” Heymach said.

Susan Galbraith, Ph.D., AstraZeneca’s oncology R&D chief, also suggested that much of the Imfinzi regimen’s benefit is going to be seen with longer-term follow-up. For now, investigators have only recorded 32% of events needed for a final EFS analysis, she noted.

On the flip side, that short follow-up time could raise some doubts about weather neoadjuvant Imfinzi is driving the EFS benefit so far, given that adjuvant Imfinzi has yet to fully kick in. But one could also argue that adjuvant Imfinzi did pull its weight because neoadjuvant Imfinzi’s pCR advantage appeared smaller than Opdivo’s, yet the two regimens eventually achieved similar EFS improvements. But its exact contribution is almost impossible to tease out.

“I do think that further maturity of this data will be important,” Galbraith said.

Now, the question is: Will the FDA think so, too?

AZ has shared the AEGEAN data with the FDA and continues to talk with the agency, Galbraith said. She declined to offer any details on progress until the FDA accepts a potential filing.

Data on patients’ overall survival remain immature, and Galbraith wouldn’t share which direction the survival trend is heading. But she did note that in CheckMate-816, neoadjuvant Opdivo only started to separate from control on overall survival at about 15 months of follow-up.

A sign for the entire PD-1/L1 class?

Still, all those explanations simply might not matter because not many doctors will dig into the trial nuances. It’s also tough to know whether Imfinzi would be much better than Opdivo even after adjusting for trial differences.

The question then is: Is it because Imfinzi is simply a weaker checkpoint inhibitor in resectable NSCLC that it needed longer, perioperative treatment to reach the same effect? In other words, can the AEGEAN trial results be extrapolated to the entire PD-1/L1 class?

Industry watchers may soon get more clarity on that question from another trial.

Days before AZ announced the AEGEAN trial success in March, Merck said the phase 3 Keynote-671 trial had hit the EFS endpoint for Keytruda in a similar neoadjuvant-plus-adjuvant regimen. Keytruda has had the most success in NSCLC, including a recent broad FDA approval as a postsurgery adjuvant treatment.

Merck has yet to share detailed results from Keynote-671. It’d be a big win for Merck if Keytruda showed an EFS improvement that’s meaningfully higher than neoadjuvant Opdivo’s 37%, which SVB Securities analyst Daina Graybosch, Ph.D., previously labeled “impressive.”

As Graybosch projected in a note a year ago, early-stage cancer treatment will become a major growth point for Keytruda, representing about a third of the drug’s total sales in 2025. Long durations of treatment in the adjuvant setting are a key driver of revenues, she said.

Despite Imfinzi and Keytruda’s trial wins, there’s still a need to understand which patients may be able to avoid further adjuvant treatment and those who might require even more intensive adjuvant regimens, Heymach said.

Besides the broader question about the overall necessity of adjuvant treatment, it’s not clear whether patients who’ve achieved a pCR after neoadjuvant immunotherapy still need to continue treatment. But neither AEGEAN nor Keynote-671 could answer that question.

For now, Heymach said that given AEGEAN’s largely consistent signal in various patient subgroups, including in PD-L1-negative patients, he would consider the Imfinzi regimen for any eligible patients under the drug’s label, and especially for patients with low PD-L1 expression who have a substantial risk of relapse.

As expected, Imfinzi’s EFS benefit was most pronounced in patients with high levels of PD-L1 expression. But the drug also chalked up a 24% risk reduction among PD-L1-negative patients.

Imfinzi showed “good activity across the levels of PD-L1 expression,” Galbraith said, “in both squamous and non-squamous patient populations, and in both combination with carboplatin and cisplatin. So all of that is important for the applicability of the treatment across the patient populations.”