Epitope detection sensitivity remains a primary bottleneck in mass spectrometry (MS)-based immunopeptidomics, as conventional discovery-based workflows such as data-dependent (DDA) and data-independent acquisition (DIA) frequently lack the sensitivity required to detect ultra-low abundant targets. While these untargeted methods are powerful for mapping the general immunopeptidome, the stochastic nature of precursor selection and the presence of complex, chimeric spectra mean that rare species—such as viral or mutation-derived neoepitopes—often remain undetected. In this study, we present optiPRM+, an ultra-sensitive targeted proteomics workflow optimized for the Orbitrap Exploris 480 platform designed to bridge this sensitivity gap through systematic, peptide-specific parameter tuning. Our approach centers on the empirical characterization of target peptides using inclusion list-driven data-dependent acquisition (iDDA) and direct infusion-MS to determine optimal fragmentation conditions. We demonstrate that substantial sensitivity gains are achieved by pushing Orbitrap scanning parameters to their practical limits, specifically utilizing ultra-high MS2 resolutions of up to 480,000, maximum ion injection times of 1000 ms, and narrow precursor isolation windows to maximize signal-to-noise ratios for trace-level analytes. Furthermore, we refined per-peptide collision energy (CE) optimization, discovering that precursors with a charge state exceeding their basic amino acid count require unusually low energies for optimal fragmentation—a finding of particular importance for the non-tryptic peptides characteristic of the immunopeptidome. We applied the optiPRM+ workflow to the challenging biological case of Human Papillomavirus type 16 (HPV16), a virus known to suppress antigen presentation pathways. This optimized strategy enabled the confident identification and validation of the HLA-A02:01-restricted epitope TIHDIILECV and the first MS-based detection of two novel viral targets: ISEYRHYCY (HLA-A01:01) and CVYCKQQLLR (HLA-A*11:01). To contextualize these findings, we performed global immunopeptidome analysis via DIA, which confirmed that these ultra-low abundance peptides were not detectable through untargeted methods despite being clearly validated by our targeted approach. By successfully detecting these viral peptides, we demonstrate that a systematically optimized targeted-first approach can uncover biologically relevant epitopes that remain invisible to conventional discovery-based workflows.