The baseline serum cytokine levels of platinum pretreated patients with advanced squamous non-small cell lung cancer (NSCLC) altered the efficacy of immunotherapy with nivolumab, a fully human IgG4 antibody inhibiting the programmed death-1 (PD-1) immune checkpoint.
Updated findings from the phase II CheckMate 063 trial of nivolumab and phase III CheckMate 017 trial of nivolumab versus docetaxel in patients withadvanced platinum-refractory squamous NSCLC were submitted by Dr Hervé Lena, Service de Pneumologie, Centre Hospitalier Universitaire de Rennes, Rennes, France and colleagues, including Suresh Ramalingam who presented the results at the European Lung Cancer Conference (ELCC), held in Geneva, Switzerland, 13 to 16 April, 2016.
Nivolumab, which targets the programmed death-1 (PD-1) receptor, was administered in CheckMate 063 at 3 mg/kg Q2W to 117 patients until progressive disease (PD) or unacceptable toxicity occurred. CheckMate 017 randomised 135 patients to receive nivolumab at the same dose and 137 patients to docetaxel at 75 mg/m2 Q3W until PD or discontinuation due to toxicity or other reasons.
The primary endpoints were overall response rate (ORR) by independent radiology review (RECIST v1.1) and overall survival (OS) in CheckMate 063 and CheckMate 017, respectively. Treatment beyond PD was permitted per protocol in both studies.
A multivariate exploratory analysis was done on the levels of baseline serum cytokines in 222 of the nivolumab-treated patients participating in either trial to generate a SQ-cytoscore, defined as “high” or “low” based on the median cut-off, which allowed quantification of the effect of the identified cytokine set on OS.
Consistent response seen with nivolumab across both trials
Similar results at a two-year follow-up were observed with nivolumab in both CheckMate trials regarding key efficacy and safety parameters.
In CheckMate 063, the 12-month OS and 18-month OS rates were 39% (95% CI 30%, 48%) and 27% (95% CI 19%, 35%) respectively, and median OS was 8.1 months (95% CI 6.1, 10.9 months). The ORR was 15% (95% CI 9%, 22%).
Overall, 75% of patients experienced a treatment-related adverse event (TRAE) of any grade and 17% of patients experienced grade 3/4 TRAEs; 12% of patients discontinued nivolumab due to a TRAE and death occurred in two patients that was determined treatment-related.
Superior outcome and safety demonstrated by nivolumab compared with docetaxel
In CheckMate 017, the 12-month OS rates of nivolumab compared to docetaxel were 42% (95% CI 33.8%, 50.4%) versus 24% (95% CI 17.4%, 31.7%) and the 18-month OS rates were 28% (95% CI 20.8%, 35.8%) versus 13% (95% CI 7.6%, 18.6%), respectively. Median OS was 9.2 months with nivolumab (95% CI 7.33, 12.62 months) versus 6.0 months with docetaxel (95% CI 5.29, 7.39 months). The ORR with nivolumab was 20% (95% CI 14%, 28%) versus 9% with docetaxel (95% CI 5%, 15%).
Of the 131 nivolumab treated patients and 129 docetaxel patients, 59% versus 87% of patients, respectively, experienced a TRAE of any grade and 8% versus 56% of patients in the respective groups experienced grade 3/4 TRAEs. Study discontinuation due to a TRAE was reported for 5% of nivolumab versus 10% of docetaxel patients. No treatment related deaths occurred with nivolumab and two deaths occurred in patients receiving docetaxel that were determined treatment-related.
Exploratory analyses from CheckMate 063 and CheckMate 017 reveal that baseline serum cytokine levels alter overall survival with nivolumab
An evaluation of baseline serum cytokines in nivolumab-treated patients showed that 102 patients having a high SQ-cytoscore achieved OS nearly 3 times longer than 120 patients having a low SQ-cytoscore; median OS was 15.6 versus 5.3 months, HR 0.48 (95% CI 0.36, 0.64) in the high and low score groups, respectively (p < 0.0001).
Monotherapy with checkpoint inhibitors: Any chance for long term survival?
Martin Reck of the Department of Thoracic Oncology, LungClinic Grosshansdorf, Germany in his discussant talk entitled ‘Monotherapy with checkpoint inhibitors: Any chance for long term survival?’ said that squamous cell NSCLC is a different and difficult disease. He underlined a number of differences in comparison to the non-squamous adenocarcinoma type of NSCLC. In terms of patient demographics, non-squamous adenocarcinoma NSCLC is the most common cell type in women and non-smokers. Patients tend to be younger, while squamous NSCLC, the most common histology in men, has a stronger association with smoking and patients tend to be slightly older.
In terms of localisation, non-squamous adenocarcinoma NSCLC is peripheral, while squamous is central (and may be more likely to result in blood vessel haemorrhages). For non-squamous adenocarcinoma cavitation is not typical, while it is typical for squamous NSCLC. In non-squamous adenocarcinoma NSCLC patients often present with metastatic disease before symptom development, while squamous NSCLC is more likely to be detected at localised stage due to earlier onset of symptoms. Brain metastases are more common in non-squamous adenocarcinoma, while squamous NSCLC commonly metastasises to regional lymph nodes, adrenal glands, bone, liver and brain. In addition, there are number of differences in term of pathology; adenocarcinoma is characterised by the presence of glands and papillary structures, neoplastic cells with round to oval nuclei, prominent nucleoli, and moderate amounts of cytoplasm, staining for mucin, TTF-1, cytokeratin 7, while squamous NSCLC by flattened appearance, intercellular bridges, individual cell keratinisation, keratin pearls, staining for p63, p40, cytokeratin 5/6 and few treatable oncogenic alterations.
Dr Reck emphasised that improvements in survival over recent decades have been greater in stage IV adenocarcinoma vs squamous NSCLC. In general, survival has been improving since 1990 for NSCLC of all histologies but significantly increased survival was observed for patients diagnosed in 2002-2005 with adenocarcinoma compared with those diagnosed with squamous cell carcinoma. However, second-line treatment for squamous cell carcinoma was not really a model for long term overall survival but it has changed recently. He reviewed the findings from studies with checkpoint inhibitors and the study presented at ELCC 2016, not only in term of efficacy, but also in terms of treatment-related adverse effects.
Furthermore, he discussed how we can characterise the ‘immunogenic’ tumour: clinical factors are not suitable and PD-L1 expression is not helpful in squamous cell carcinoma. However, patients with high SQ-cytoscore showed longer overall survival (prognostic factor). But he questioned associating dynamic changes in cytokine levels with the efficacy of checkpoint inhibitors. Besides cytokines, could circulating immune cells also serve as a marker of interest, and do we see similar outcomes also in non-squamous NSCLC? Analyses of CM-057 will be presented at ASCO 2016. In terms of the question: is there is a chance for long term survival with checkpoint-inhibitors? Dr Reck concluded yes, it seems so, but we are still on the way to identifying these patients.
The authors noted that patients with platinum refractory, advanced squamous NSCLC had few treatment options until nivolumab was approved for these patients in both the United States and the European Union. Nivolumab approval was based on a survival advantage over docetaxel that was demonstrated in CheckMate 063 and CheckMate 017.
The results presented here today confirm the efficacy of nivolumab in patients with advanced squamous NSCLC and demonstrate that this efficacy is impacted by the cytokine level in patients prior to treatment, suggesting that baseline serum cytokine levels may also serve as a marker for nivolumab efficacy.
CheckMate 063 and CheckMate 017 were funded by Bristol-Meyers Squibb