New data on PD-L1 inhibitor activity and determinants of outcomes in immunotherapy-treated patients

Superior survival with durvalumab in the PACIFIC trial

Traditionally, the standard-of-care treatment for patients with unresectable, stage III non–small-cell lung cancer (NSCLC) used to be platinum-based chemoradiotherapy. However, outcomes have been poor, which provided the rationale for the phase III PACIFIC trial. PACIFIC investigated the anti-PD-L1 antibody durvalumab 10 mg/kg every 2 weeks (Q2W) for up to 12 months (n = 476) versus placebo (n = 237) in patients who had remained progression-free after definitive platinum-based concurrent chemoradiotherapy. These patients had been recruited irrespective of their PD-L1 status. The superiority of durvalumab concerning progression-free survival (PFS) was shown at the time of the first planned interim analysis, which yielded a PFS improvement of 11.2 months [1].
At the WCLC 2018, Antonia et al. reported the second primary endpoint of overall survival (OS) as well as updated results for PFS and other secondary endpoints [2]. For OS, durvalumab showed statistically significant and clinically meaningful improvement over placebo in the intent-to-treat (ITT) population. While the median OS had not yet been reached in the experimental arm, it was 28.7 months in the placebo group (HR, 0.68; p = 0.00251; Figure 1). At 24 months, 66.3 % vs. 55.6 % of patients were alive. As in the first interim analysis, the PFS difference in favour of durvalumab amounted to 11 months (17.2 vs. 5.6 months; HR, 0.51). Likewise, improvements versus placebo with respect to time to death or distant metastasis as well as incidence of new lesions were maintained.
Both OS and PFS results favoured durvalumab across a range of patient subgroups. Two PD-L1 analyses were performed, with one being pre-specified and the other one being an unplanned post-hoc analysis. The cut points differed between the analyses (25 % and 1 %, respectively). Patients whose tumours were PD-L1-negative according to the post-hoc analysis did not benefit from durvalumab treatment with regard to OS or PFS. No new safety signals were identified after longer follow-up. The authors noted that PACIFIC is the first study to demonstrate a survival advantage for unresectable, stage III NSCLC, supporting chemoradiotherapy followed by durvalumab for 1 year as a standard of care.

Figure 1: Significant reduction of mortality risk with durvalumab compared to placebo in patients with unresectable, stage III NSCLC after chemoradiotherapy

Figure 1: Significant reduction of mortality risk with durvalumab compared to placebo in patients with unresectable, stage III NSCLC after chemoradiotherapy

IMpower133: atezolizumab plus SCLC standard treatment

There has been little progress in the first-line management of patients with small-cell lung cancer (SCLC) for more than 20 years. The majority of patients present with extensive-stage SCLC (ES-SCLC); here, the standard of care is platinum plus etoposide. In spite of high initial response rates, outcomes remain poor.
The global, phase I/III, double-blind, randomised, placebo-controlled IMpower133 study evaluated first-line treatment with the anti-PD-L1 antibody atezolizumab plus carboplatin and etoposide (n = 201) compared to placebo plus carboplatin and etoposide (n = 202) in patients with ES-SCLC [3]. Indeed, IMpower133 was the first study in more than 20 years to show a clinically meaningful improvement in OS over the current first-line standard of care. The addition of atezolizumab significantly prolonged both OS (12.3 vs. 10.3 months; HR, 0.70; p = 0.0069) and investigator-assessed PFS (5.2 vs. 4.3 months; HR, 0.77; p = 0.017) that were defined as the co-primary endpoints. At 12 months, 51.7 % vs. 38.2 % of patients were alive, and 12.6 % vs. 5.4 % were progression-free. Except for patients with brain metastases, the OS subgroup analysis favoured the atezolizumab-based regimen across all of the subgroups, which means that the benefit of treatment was not limited to patients with high tumour mutational burden (TMB) (Figure 2). Response rates did not differ across the two arms, although duration of response favoured the experimental arm (4.2 vs. 3.9 months; HR, 0.70). The atezolizumab combination also gave rise to greatly improved results with respect to event-free rates at 6 months (32.2 % vs. 17.1 %) and 12 months (14.9 % vs. 6.2 %). A greater proportion of patients in the experimental arm had ongoing responses (14.9 % vs. 5.4 %). Possible correlations of survival outcomes with the PD-L1 status have not been established yet, but will be assessed in the future.
Rates of AEs were similar across the two arms. Haematological toxicity occurred most commonly in the entire population, with the addition of atezolizumab not altering the incidence. The median number of carboplatin and etoposide doses received was identical in the two groups, which implies that ­atezolizumab treatment did not interfere with dose delivery. No new safety signals were identified. Overall, these data suggest that atezolizumab plus carbo­platin and etoposide is a new standard-of-care first-line treatment for patients with ES-SCLC.

Figure 2: IMpower133: overall survival benefits in all subgroups including those with low tumour mutational burden, except for patients with brain metastases

Figure 2: IMpower133: overall survival benefits in all subgroups including those with low tumour mutational burden, except for patients with brain metastases

Promising activity of neoadjuvant atezolizumab

The multicentre, open-label, single-arm, phase II LCMC3 study tested the neoadjuvant use of atezolizumab in patients with resectable NSCLC. Preliminary efficacy and safety data obtained in Part 1 of the trial have been reported at the ASCO 2018 Congress [4].
At the WCLC 2018, Rusch et al. presented updated safety and efficacy results in 54 patients [5]. Forty-five individuals without EGFR or ALK aberrations underwent surgical resection. Ten of these (22 %) achieved major pathological responses, which were defined as ≤ 10 % viable tumour cells. Three patients (7 %) had pathological complete remissions. Changes in lesion size from baseline appeared not to correlate with the amount of viable tumour cells. The neoadjuvant administration of atezolizumab proved tolerable and did not cause any major delays in surgery or interfere with the surgical resection. A follow-up interim analysis in 90 patients is planned.

Second-line avelumab: negative trial according to primary analysis

No OS benefit was detected in the randomised, open-label, phase III JAVELIN Lung 200 trial that tested the anti-PD-L1 antibody avelumab 10 mg/kg Q2W against docetaxel in the pre-treated setting [6]. These patients had experienced disease progression after platinum doublet therapy. The primary analysis population comprised 529 patients who showed PD-L1 expression levels ≥ 1 %.
In this group, there was no difference regarding the primary endpoint of OS (11.4 vs. 10.3 months; HR, 0.90; p = 0.1627); however, as the authors pointed out, OS findings in this trial might have been impacted by high use of subsequent immune checkpoint inhibitor (ICI) treatment in the docetaxel arm. Prespecified exploratory subgroup analyses showed increasing clinical activity of avelumab compared to docetaxel in patients with higher PD-L1 expression; for instance, in those with ≥ 80 % expression, median OS was 17.1 vs. 9.3 months (HR, 0.59; p = 0.0022). This group also experienced the greatest benefits with regard to PFS (5.6 vs. 2.8 months; HR, 0.58; p = 0.0021) and objective response rate (ORR; 31 % vs. 10 %; p = 0.0002). Avelumab had an overall favourable AE profile relative to docetaxel. Several trials assessing avelumab in NSCLC patients are ongoing, including JAVELIN Lung 100, which is a phase III trial of first-line avelumab monotherapy for PD-L1-positive NSCLC (NCT02576574).

Effect of antibiotic treatment on survival endpoints

Retrospective data suggest that the use of antibiotics alters patient response to ICIs in different types of cancer including NSCLC [7, 8]. This may be due to the fact that antibiotics affect the gut microbiota, which plays an essential role in the development and maturation of the immune system [9].
A multicentre, retrospective study conducted in 168 consecutive NSCLC patients who received nivolumab or pembrolizumab in the second line or beyond revealed that the use of antibiotics appeared to have a negative impact on survival outcomes [10]. Antibiotics were administered 2 months before or within the first month after the beginning of ICI treatment in almost half of the patients. Median OS was significantly shorter in patients who received antibiotics compared to those who did not (8.1 vs. 11.9 months; HR, 1.55; p = 0.027).
The investigators also determined if OS was affected by the route of administration of antibiotics. They found that patients treated intravenously fared markedly worse than those on oral treatment (HRs, 3.62 and 1.17, respectively). The multivariate analysis confirmed intravenous application as an independent risk factor. Consistent results were obtained with regard to both antibiotic use and route of administration for the endpoint of PFS.
Moreover, the type of infection might also affect the outcomes, as patients with lower respiratory tract and urinary infections experienced considerably shorter median OS than those with other infections (6 vs. 26 months; p = 0.006). In light of the retrospective nature of this analysis, the results need further prospective confirmation. For the time being, the authors recommended a rational use of antibiotics in ICI-treated patients.

Driver mutations and outcomes with ICI therapy

ICIs are thought to be less effective in patients whose lung tumours harbour oncogenic driver mutations, but data are limited due to low mutation frequency and exclusion of these patients from clinical trials. Vokes et al. therefore assessed clinical outcomes in 82 ICI-treated patients with targetable driver mutations, including EGFR aberrations (L858R mutation, exon 19 deletion, exon 20 insertion, missense mutation in OncoKB), ALK, ROS1 and RET rearrangements, MET exon 14 skipping mutations (METΔ14), and BRAF V600E missense mutations [11]. These patients received immunotherapy in the third or later lines. The TMB was calculated as the number of non-synonymous mutations per megabase of genome covered.
PFS in this group of patients did not differ from PFS in a wild-type population. Considering the low numbers, however, a trend was observed towards worse outcomes in patients with EGFR, RET, ROS1 and ALK aberrations (Table). Conversely, PFS in patients with BRAF and METΔ14 appeared similar to that obtained in the wild-type setting. Except for the group with RET-mutant tumours, at least one patient in each subtype group achieved PFS of > 6 months. PFS results of > 12 months occurred in at least one individual with EGFR, BRAF or METΔ14 aberrations, which indicates that some of these patients derived significant clinical benefit from ICI treatment. Likewise, response rates did not differ significantly between patients with driver mutations and those without, although a trend was observed towards lower response rates in patients with EGFR, ALK and RET aberrations. An exploratory analysis according to EGFR-mutant subtypes demonstrated that fewer patients in the groups with L858R mutation and exon 20 insertion responded compared to those in the other groups. As expected, TMB was lower in the population with driver mutations than in the wild-type population. This marker did not correlate with response in the group with driver mutations, neither in the entire cohort nor in different mutation subtypes.
The investigators concluded that even though response rates may be lower in certain mutation subtypes, neither the presence of oncogenic driver mutations nor low TMB should preclude offering these patients therapeutic trials of ICI therapy. Further retrospective and prospective studies are necessary.

Progression-free survival in patients treated with immunotherapies according to the type of oncogenic driver aberration


  1. Antonia SJ et al., Durvalumab after chemoradiotherapy in stage III non-small-cell lung cancer. N Engl J Med 2017; 377(20): 1919-1929
  2. Antonia SJ et al., Overall survival with durvalumab versus placebo after chemoradiotherapy in stage III NSCLC: Updated results from PACIFIC. WCLC 2018, PL02.01
  3. Liu SV et al., IMpower133: primary PFS, OS, and safety in a ph1/3 study of 1L atezolizumab + carboplatin + etoposide in extensive-stage SCLC. WCLC 2018, PL02.07
  4. Rusch VW et al., Neoadjuvant atezolizumab in resectable non-small cell lung cancer (NSCLC): Initial results from a multicenter study (LCMC3). J Clin Oncol 36, 2018 (suppl; abstr 8541)
  5. Rusch VW et al., Neoadjuvant atezolizumab in resectable non-small cell cancer (NSCLC): updated results from a multicentre study (LCMC3). WCLC 2018, MA04.09
  6. Barlesi F et al., Avelumab vs docetaxel for previously treated advanced NSCLC: primary analysis of the phase 3 JAVELIN Lung 200 trial. WCLC 2018, OA05.05
  7. Derosa L et al., Negative association of antibiotics on clinical activity of immune checkpoint inhibitors in patients with advanced renal cell and non-small-cell lung cancer. Ann Oncol 2018; 29(6): 1437-1444
  8. Routy B et al., Gut microbiome influences efficacy of PD-1-based immunotherapy against epithelial tumors. Science 2018; 359(6371): 91-97
  9. Ubeda C & Pamer EG, Antibiotics, microbiota, and immune defense. Trends Immunol 2012; 33(9): 459-466
  10. Mielgo Rubio X et al., Antibiotic use and PD-1 inhibitors: shorter survival in lung cancer, especially when given intravenously. Type of infection also matters. WCLC 2018, MA10.01
  11. Vokes N et al., Efficacy and genomic correlates of response of anti-PD1/PD-L1 blockade in non-small cell lung cancers harboring targetable oncogenes. WCLC 2018, MA19.01