In this paper, we revised the current understanding of the protein corona that is created on the surface of nanoparticles in blood plasma after an intravenous injection. We have focused on nanoparticles that have a proven therapeutic outcome. These nanoparticles are based on two types of biocompatible amphiphilic copolymers based on N-(2-hydroxypropyl)methacrylamide (HPMA): a block copolymer, poly(ε-caprolactone) (PCL)-b-poly(HPMA), and a statistical HPMA copolymer bearing cholesterol moieties, which have been tested both in vitro and in vivo. We studied the interaction of nanoparticles with blood plasma and selected blood plasma proteins by electron paramagnetic resonance (EPR), isothermal titration calorimetry, dynamic light scattering, and cryo-transmission electron microscopy. The copolymers were labeled with TEMPO radicals at the end of hydrophobic PCL or along the hydrophilic HPMA chains to monitor changes in polymer chain dynamics caused by protein adsorption. By EPR and other methods, we were able to probe specific interactions between nanoparticles and blood proteins, specifically low- and high-density lipoproteins, immunoglobulin G, human serum albumin (HSA), and human plasma. It was found that individual proteins and plasma have very low binding affinity to nanoparticles. We observed no hard corona around HPMA-based nanoparticles; with the exception of HSA the proteins showed no detectable binding to the nanoparticles. Our study confirms that a classical "hard corona-soft corona" paradigm is not valid for all types of nanoparticles and each system has a unique protein corona that is determined by the nature of the NP material.