Both IMpower nab paclitaxel carboplatin plus atezolizumab an
Both IMpower-130 (nab-paclitaxel/carboplatin plus atezolizumab) and IMpower-150 (atezolizumab/carboplatin/paclitaxel plus bevacizumab) enrolled patients with EGFR and ALK positive tumors who had failed at least one line of appropriate TKI therapy. In the subgroup analyses there was a trend for improved OS for EGFR mutant and ALK positive patients treated with the bevacizumab-containing regimen in IMpower-150 (HR for EGFR/ALK positive group: 0.54; 95% CI, 0.29–1.03, HR for patients with a sensitizing EGFR mutation: 0.39, 95% CI, 0.14–1.07). This potential benefit in OS was not seen without bevacizumab in IMpower-130 (HR, 0.98; 95% CI, 0.41–2.31). Studies that investigated single agent IO in patients with activating EGFR mutation clearly demonstrated a lack of OS benefit in this population even in patients with high PD-L1 expression [33,34]. The potential signal of activity in EGFR or ALK positive patients seen in IMpower-150 would need prospective validation and due to the exploratory nature and the small sample sizes of the analysed subgroups these data must be interpreted with caution. The synergism of bevacizumab with atezolizumab seems also to be relevant for patients with liver metastases. In IMpower-150, the addition of bevacizumab to atezolizumab plus carboplatin/paclitaxel revealed an OS benefit in patients with liver metastases (HR 0.52; 95% CI 0.33–0.82) . This benefit was not reported in the IMpower-130, -131 and -132, which investigated regimens without bevacizumab. The liver immune environment is known to contain immune-suppressive myeloid cells expressing VEGFR2, a fact that would explain the ability of bevacizumab to overcome this immune suppression in decreasing these cells and leading to an improved efficacy of a immune CPI in the liver [35,36]. These subgroup analyses also have to be interpreted with caution due to their retrospective character and the small sample sizes. A retrospective exploratory subgroup analysis of the KEYNOTE-189 trial, presented at the 2019 Annual American Association of Cancer Research (AACR) meeting, evaluated survival outcomes in patients with presence or absence of liver or Gilteritinib metastases at baseline. 115 of 616 patients (18%) presented with baseline liver metastases. The pembrolizumab plus pemetrexed/platinum combination showed an OS improvement from 6.6 to 12.6 months (HR 0.62, 95% CI, 0.39-0.98) in patients with liver metastases and from 13.2 to 23.7 months (HR 0.58, 95% CI, 0.45-0.74) in patients without liver metastases. . 108 of 616 patients (17.5%) presented with stable baseline brain metastases. The pembrolizumab plus pemetrexed/platinum combination showed an OS improvement from 7.5 to 19.2 months (HR 0.41, 95% CI, 0.24-0.67) in patients with brain metastases, and an improvement from 12.1 to 22.4 (HR 0.59, 95% CI, 0.46-0.75) for patients without brain metastases. Similarly to the overall population, the risk of death was reduced by approximately a half in the IO/ChT combination irrespective of the presence or absence of liver or brain metastases and no new safety signals were identified. On the basis of available data, it is challenging to suggest the best treatment approach for patients with pre-existing brain metastases. All IO-based first-line registration studies only included patients with treated and stable brain metastases. As bevacizumab demonstrated encouraging efficacy and acceptable safety with first-line paclitaxel and carboplatin in patients with NSCLC and asymptomatic, untreated brain metastases in the BRAIN study  a bevacizumab containing IO based regimen might possibly be a safe and effective option in this setting. Innovative predictive biomarkers, beyond PD-L1 expression, are urgently needed to improve patient selection and treatment outcomes. TMB is an emerging predictor of response. A retrospective unplanned exploratory analysis assessed the impact of TMB on treatment with nivolumab in the CheckMate-026 study. A high TMB (≥243 somatic mutations assessed by whole-genome-sequencing) indicated a trend towards longer PFS (HR, 0.62; 95% CI, 0.38–1.00) and higher ORR in favour of nivolumab. In the same analysis, the three TMB strata (high, intermediate and low) occurred to be independently distributed, irrespective of PD-L1 expression . In the CheckMate-227 study, a high TMB lead to longer PFS in patients treated with ipilimumab plus nivolumab compared to ChT. A recently published analysis investigated the correlation between TMB (assessed by whole-genome-sequencing) and response to ipilimumab plus nivolumab in 75 NSCLC patients treated in the phase I CheckMate-012 study . TMB was found to be a strong predictor of efficacy for ipilimumab plus nivolumab with significantly higher TMB in patients with a complete or partial response (CR/PR) compared to patients with stable disease of progressive disease (SD/PD) (P = 0.0004). Furthermore, high TMB was associated with a durable clinical benefit (PR or SD > 6 months). No correlation between TMB and PD-L1 expression was detected, once again highlighting TMB as an independent predictor for immune therapy . However, in the CheckMate-026 the PFS-benefit did not translate into an OS benefit and OS data of CheckMate-227 are still awaited. Notably, in both studies the TMB assessment was done retrospectively and further prospective analyses are needed. Another potential predictive biomarker is T-effector-signature-expression, which was prospectively evaluated in IMpower-150. It is defined as messenger RNA expression of three different genes (PD-L1, CXCL9, INF-y) as a surrogate for PD-L1 expression and was measured by real-time PCR on tumor tissue at baseline. In the OAK trial, T-effector gene signature expression occurred to be a more sensitive biomarker for a PFS benefit than PD-L1 expression . In the IMpower-150 study, patients with a high T-effector gene signature-expression (Teff-high) had a longer PFS (11.3 vs. 6.8 months, HR, 0.51; 95% CI, 0.38- 0.86; P < 0.001) when treated with ABCP compared to BCP and also a trend toward longer OS (25.0 vs. 16.7 months, HR, 0.83; 95% CI, 0.59–1.17) . Interferon-γ messenger RNA gene expression, tumor infiltrating lymphocytes (TILS) and tumor environmental infiltrating lymphocytes are also promising predictors under investigation [, , ].