Circulating DNA (ctDNA) and specifically the detection cancer-associated mutations in liquid biopsies promises to revolutionize cancer detection. The main difficulty however is that the length of typical ctDNA fragments (~150 bases) can form secondary structures potentially obscuring the mutated fragment from detection. We show that an assay based on gold nanoparticles (65 nm) stabilized with DNA (Au@DNA) can discriminate single nucleotide polymorphism in clinically-relevant ssDNA sequences (70-140 bases). The preincubation step was crucial to this process, allowing sequential bridging of Au@DNA, so that single base mutation can be discriminated, down to 100 pM concentration.Detection of single nucleotide polymorphisms (SNP) in ssDNA sequences by selective aggregation of plasmonic nanoparticles carries the potential for rapid determination of cancer biomarkers in liquid biopsies such as blood.1-3 Although, the detection of sequences of up to 40 bases without any signal amplification has been demonstrated 4,5 the average size of circulating DNA is ~150 bases in length, 6 and these longer fragments are more problematic as the increase of the sequence length affects the plasmon coupling (larger gaps between the particles), that decrease the limit of detection. 7 In addition, long DNA sequences form thermodynamically stable secondary structures that render the detection even more difficult.8 Therefore, there is an obvious need for new solutions to detect long DNA sequences.In earlier studies, we have reported a plasmonic assay comprising DNA-coated gold nanoparticles (AuNPs) that could discriminate SNP in less than 10 min. 9 Even though relatively large particles may lead to plasmon coupling upon aggregation, the assay is limited to the detection of rather short sequences (up to 23 bases). We report here that large plasmonic particles (65 nm) functionalized with modified DNA 10 can be used to detect single-base mutation in clinically relevant sequences with 140 bases in length. We found out that premixing the AuNPs (batch 1) with the target DNA was sufficient to induce aggregation upon addition of batch 2 (Figure 1b). In contrast, the addition of the target DNA to the mixture of two batches (standard sandwich assay, Figure 1a) had no effect on the aggregation.As a target we choose the most common point mutation in non-small cell lung cancer (NSCLC), the L858R mutation that occurs in the Epidermal Growth Factor Receptor (EGFR) gene.11 Detection of this mutation is an FDA and EMA-approved biomarker for the administration of TKItherapy to NSCLC patients. Figure 1d-e show possible secondary structures of target sequences comprising 70 and 140 bases, with indication of the binding sites of Au@DNA and the location of a single-point mutation. As a signal transducer, we used AuNPs with 65 nm in diameter, stabilized by short sequences that were complementary to either a mutation-free region in the target DNA (WT) or to a region that contained a single-base mutation (MUT) (Figure 1c). Each AuNP was stabilised by ~15...