The melanoma-associated antigen 3 (MAGE-A3) is found in approximately 35 % of resectable NSCLC. This antigen is not expressed in normal cells beyond embryogenesis [9]. Thus, it is a suitable target for vaccination.

In a double-blind, placebo controlled phase II trial, 122 patients received recombinant MAGE-A3 combined with an immunostimulant and 60 received placebo following resection of NSCLC stage Ib or II. There were 13 doses given in a 27-month period. Primary end point was the disease-free interval (DFI). In the 44 months observation period after resection, 35 % of relapses were seen in the MAGE-A3 arm compared to 43 % in the placebo arm. However, there was no statistical significant difference regarding DFI (hazard ratio [HR], 0.75; 95 % CI, 0.46–1.23; two-sided p 0.254), disease-free survival (DFS; HR, 0.76; 95 % CI, 0.48–1.21; p 0.248), or overall survival (OS; HR, 0.81; 95 % CI, 0.47–1.40; p 0.454). In all patients, a humoral immune response was detected and no relevant toxicity was observed [10].

In an accompanying project, a gene signature was developed to predict clinical benefit from the vaccine: Patients from two previous phase II studies in metastatic and resected melanoma, respectively, served as the testing cohort. In the patients with stage IV melanoma, MAGE-A3 was given with two different adjuvants, named AS15 and AS02 [11]. The resected patients received either placebo or MAGE-A with AS02 as adjuvants [10]. In the patients with melanoma, the expression of 84 genes was associated with clinical benefit. A clinical significant effect was only seen with AS15 (HR for overall survival, 0.37; 95 % CI, 0.13–1.05; p 0.06) but not with AS02 (HR, 0.84; 95 % CI, 0.36–1.97; p 0.70). These results were then validated using a cohort of resected NSCLC patients, focussing on those receiving either MAGE-A3 plus AS15 or placebo. A benefit from the vaccine was only seen in gene signature-positive patients (HR, 0.42; 95 % CI, 0.17–1.03; p 0.06). In patients classified as gene signature negative, there was no difference between active treatment and placebo (HR, 1.17; 95 % CI, 0.59–2.31; p 0.65) [12].

A phase III study (MAGRIT; clinical trial number: NCT 00480025) with 2278 MAGE-positive patients with resected NSCLC stage IB to IIIa has completed recruitment. In this trial, patients received either MAGE-A3 with the adjuvants CpG7909 or placebo. In this trial, the above-mentioned 84-gene signature is tested for its predictive capability; however, a very recent press release announced futility of the trial in terms of the end points.

PRAME (preferentially expressed antigen in melanoma) is an antigen is found in acute leukaemias and several solid tumors like melanomas and NSCLC and an expression pattern similar to MAGE-A3, but presumably more frequently found in NSCLC than MAGE. It is not expressed in normal tissues except testis. In a pilot study, PRAME induces a cytotoxic T-cell response in 8 of 14 patients [13]. To date, a randomised phase II trial patients with resected stage I to IIIa NSCLC receive either PRAME plus AS15 adjuvant system or placebo. Primary end point is DFS (clinical trial number: NCT01853878).

Mucin-1 is an antigen found in all tissues of the human body. Its physiological function is the avoidance of cell adherence, i.e. with pathogens. It is overexpressed in a variety of malignancies like adenocarcinomas, lymphomas and myelomas. By this overexpression, cancer cells lose their polarity and MUC-1 is also expressed at the basolateral aspect of the cell, thereby reducing the cell-to-cell adhesion. This is presumably an early step in the development of metastasis [14]. Mucin-1 is also known as CA15-3, a prognostic tumour marker for surveillance in breast cancer [15]. Mucin-1 in tumour cells differs from its physiological variant in terms of an aberrant glycosylation pattern, stabilising the molecule [16]. Tecemotide (Stimuvax^®) is a liposomal vaccine against Mucin-1. It was tested in a phase II trial in patients with advanced (stage IIIB and IV) NSCLC. 171 patients who were nonprogressive after first-line chemotherapy were randomised to vaccination versus best supportive care. Patients in the tecemotide group received a single infusion of cyclophosphamide (300 mg/m²) followed by 8 weekly injections and subsequently injections every 6 weeks of the vaccine. Primary end point was overall survival. There was a positive trend favouring the vaccine with 4.4 months survival prolongation (HR = 0.739; 95 % CI, 0.509–1.073; p = .112). The effect was more pronounced in the subgroup of locally advanced patients (adjusted HR = 0.524; 95 % CI, 0.261–1.052; p = .069) [17] triggering the subsequent phase III trial in this population, the START Trial: 1513 patients with unresectable stage III NSCLC were enrolled. Patients had to be at least with stable disease after chemoradiotherapy (either concurrent or sequential) with platinum-based chemotherapy and radiation > =50 Gy. These patients were randomised 2:1 to vaccination or placebo. The drugs were administered weekly for eight times followed by administration every 6 weeks until disease progression. Again, 300 mg/m² cyclophosphamide or placebo was given 3 days prior to Stimuvax and placebo treatment, respectively. Off note, due to safety concerns, the study was halted for a median of 135 days until additional safety data was available. The patients most likely affected by this measure (those recruited < =6 months before hold) were excluded. Two hundred seventy-four patients were thus excluded from the analysis. Primary end point was overall survival. Sixty-five percent of patients received concurrent chemoradiotherapy with the remaining 35 % sequential radiotherapy after chemotherapy. Median OS was 25.6 months with tecemotide versus 22.3 months with placebo (adjusted HR 0.88, 95 % CI 0.75–1.03, p = 0.123). Focussing on the predefined subgroup of those patients who had concurrent chemoradiotherapy, the results became significant with an overall survival of 30.8 months compared to 20.6 months in the placebo arm (HR 0.78, 95 % CI 0.64–0.95, p = 0.016). There was no excess toxicity observed compared to placebo. The reason for the favourable outcome in the concurrent arm is not clear yet. Several explanations come to mind: First, it may be an issue of patient selection. Those in the concurrent arm were twice as likely being properly staged with either positron-emission tomography (PET) or mediastinoscopy and were more likely being treated with a curative intent. Second, in the concurrent arm, taxanes (i.e. paclitaxel) were more frequently used in the chemotherapy protocols than in the sequential arm. This drug may have immune-stimulating effects [18]. In conclusion, further studies are warranted because of the relevant benefit in this patient group.

TG4010 is based on a recombinant Vaccinia virus that expresses the Mucin-1 antigen and interleukin-2 (IL-2). The latter shall boost the T-cell response. In a phase IIb trial, 148 patients with unresectable stage IIIb or IV disease who expressed Mucin-1 by immunohistochemistry were randomised to chemotherapy (cisplatin and gemcitabine) plus TG4010, given weekly for six times and every 3 weeks thereafter until disease progression or to chemotherapy alone. Primary end point was the progression-free survival (PFS) fraction after 6 months. 6-month PFS was 43.2 % (32 of 74; 95 % CI 33.4–53.5) in the TG4010 plus chemotherapy group, and 35.1 % (26 of 74; 25.9–45.3) in the chemotherapy alone group.

Injection site pain (5.5 % versus 0 %) was only observed in the TG4010 group; fever (23.3 % versus 8.3 %) and abdominal pain (16.4 % versus 2.8 %) were more frequent in the TG4010 group. Anorexia (4.1 % versus 13.9 %) and pleural effusion (0 % versus 5.6 %) were the only severe adverse events in significantly different magnitude, both favouring the TG4010-Arm [19].

Belagenpumatucel (Lucanix^®) is a cryo-conserved suspension of four different NSCLC cell lines. Although this is a very specific vaccine, it is not very immunogenic. Therefore, antisense-TGF-ß DNA is added as adjuvants. Elevated levels of TGF-ß are associated with enhanced immunosuppression in patients with lung cancer. The antisense DNA thus blocks the formation of TGF-ß and thereby enhances immunogenity.

In a phase II trial, patients with stages II, IIIa, IIIb and IV received one of three doses of the vaccine (1.25, 2.5 or 5.0 × 10⁷ cells/injection) on a monthly or every-other-month schedule to a maximum of 16 injections. Immune function, safety and anticancer activity were monitored. Seventy-five patients were included, 61 in stages IIIB and IV. In these advanced stages, a 15 % response rate was achieved. In those patients receiving 2.5 or 5.0 × 10⁷ cells, the 1- and 2-year survival rates were 68 and 52 %, respectively, across all stages. In the 2.5 × 10⁷ group, the values were 39 and 20 %, respectively. In the 61 patients with advanced disease, those with a response showed significantly increased levels of cytokines compared to those patients with progressive disease (interferon gamma, p = .006; interleukin [IL]-6, p = .004; IL-4, p = .007) [20].

In a second phase II trial, patients received 2.5 × 10⁷ cells/injection monthly for 16 months. Besides clinical efficacy, the levels of circulating tumour cells (CTC) were measured to test the hypothesis that their number correlates with survival. Twenty-one patients with advanced NSCLC were enrolled on this study. No significant toxic effect was observed. Overall survival was 562 days. The median survival was 660 days in patients having less than 2 CTCs at baseline compared to 150 days in patients with 2 or more CTCs (p = 0.025) [21].

A randomised phase III trial was launched including patients with stage IIIA (T3N2 only) and IIIB and IV patients who had prior platinum-based chemotherapy with or without radiotherapy. Patients received Lucanix^® or placebo for 18 months every 4 weeks with two additional injections at month 21 and 24 in the absence of toxicity. Primary end point is overall survival. Recruitment has completed but results are pending (NCT00676507).

Epidermal growth factor (EGF) is overexpressed in many NSCLC; thus, vaccination against this target seems reasonable [22]. CIMAvax EGF^® is a recombinant human EGF receptor coupled with a transporter protein, p64k – which is derived from Neisseria meningitidis – and an adjuvant. It is used for approximately 18 years in Cuba. An analysis of a phase II trial with 40 patients and 40 patients taken from a running phase III trial regarding the immune response was published in 2011. In the phase II trial, patients received a single-site injection (0.6 mg of antigen) every 3 months followed by monthly injections, whereas in the phase III trial, the patients received 4 injections (2.4 mg of antigen) at different sites in the same time schedule. The authors classified the patients as “good antibody responders” if they achieved an anti-EGF antibody titre of ≥1:4,000 (52.8 % of patients in the phase II trial and 56.4 % in the phase III trial, respectively) and “supergood antibody responders” if they achieved an antibody titre of ≥1:64,000 (10.8 and 30.8 % of patients, respectively). Overall survival was significantly longer in the phase III trial (26.87 months versus 11.76 months in the group with good antibody responders; p < 0.05) [23]. There were no stage-adapted results reported and no results from the placebo group of the phase III trial. So, for further conclusions, we have to wait for the results of the phase III trials. Except for the one reported, there are two phase III trials conducted in the United Kingdom (NCT01444118) and Malaysia (NCT00516685). Both completed recruitment and no results are published so far.