The development of a quantitative understanding of viral evolution and the fitness landscape in HIV-1 drug resistance is a formidable challenge given the large number of available drugs and drug resistance mutations. We analyzed a dataset measuring the in vitro fitness of 70,081 virus samples isolated from HIV-1 subtype B infected individuals undergoing routine drug resistance testing. We assayed virus samples for in vitro replicative capacity in the absence of drugs as well as in the presence of 15 individual drugs. We employed a generalized kernel ridge regression to estimate main fitness effects and epistatic interactions of 1,859 single amino acid variants found within the HIV-1 protease and reverse transcriptase sequences. Models including epistatic interactions predict an average of 54.8% of the variance in replicative capacity across the 16 different environments and substantially outperform models based on main fitness effects only. We find that the fitness landscape of HIV-1 protease and reverse transcriptase is characterized by strong epistasis.

Although fitness landscapes are central to evolutionary theory, so far no biologically realistic examples for large-scale fitness landscapes have been described. Most currently available biological examples are restricted to very few loci or alleles and therefore do not capture the high dimensionality characteristic of real fitness landscapes. Here we analyze large-scale fitness landscapes that are based on predictive models for in vitro replicative fitness of HIV-1. We find that these landscapes are characterized by large correlation lengths, considerable neutrality, and high ruggedness and that these properties depend only weakly on whether fitness is measured in the absence or presence of different antiretrovirals. Accordingly, adaptive processes on these landscapes depend sensitively on the initial conditions. While the relative extent to which mutations affect fitness on their own (main effects) or in combination with other mutations (epistasis) is a strong determinant of these properties, the fitness landscape of HIV-1 is considerably less rugged, less neutral, and more correlated than expected from the distribution of main effects and epistatic interactions alone. Overall this study confirms theoretical conjectures about the complexity of biological fitness landscapes and the importance of the high dimensionality of the genetic space in which adaptation takes place.

HIV-1 replicative capacity (RC) provides a measure of within-host fitness and is determined in the context of phenotypic drug resistance testing. However it is unclear how these in-vitro measurements relate to in-vivo processes. Here we assess RCs in a clinical setting by combining a previously published machine-learning tool, which predicts RC values from partial pol sequences with genotypic and clinical data from the Swiss HIV Cohort Study. The machine-learning tool is based on a training set consisting of 65000 RC measurements paired with their corresponding partial pol sequences. We find that predicted RC values (pRCs) correlate significantly with the virus load measured in 2073 infected but drug naïve individuals. Furthermore, we find that, for 53 pairs of sequences, each pair sampled in the same infected individual, the pRC was significantly higher for the sequence sampled later in the infection and that the increase in pRC was also significantly correlated with the increase in plasma viral load and with the length of the time-interval between the sampling points. These findings indicate that selection within a patient favors the evolution of higher replicative capacities and that these in-vitro fitness measures are indicative of in-vivo HIV virus load.

Trastuzumab is effective in the treatment of HER2/neu over-expressing breast cancer, but not all patients benefit from it. In vitro data suggest a role for HER3 in the initiation of signaling activity involving the AKT–mTOR pathway leading to trastuzumab insensitivity. We sought to investigate the potential of HER3 alone and in the context of p95HER2 (p95), a trastuzumab resistance marker, as biomarkers of trastuzumab escape. Using the VeraTag® assay platform, we developed a dual antibody proximity-based assay for the precise quantitation of HER3 total protein (H3T) from formalin-fixed paraffin-embedded (FFPE) breast tumors. We then measured H3T in 89 patients with metastatic breast cancer treated with trastuzumab-based therapy, and correlated the results with progression-free survival and overall survival using Kaplan–Meier and decision tree analyses that also included HER2 total (H2T) and p95 expression levels. Within the sub-population of patients that over-expressed HER2, high levels of HER3 and/or p95 protein expression were significantly associated with poor clinical outcomes on trastuzumab-based therapy. Based on quantitative H3T, p95, and H2T measurements, multiple subtypes of HER2-positive breast cancer were identified that differ in their outcome following trastuzumab therapy. These data suggest that HER3 and p95 are informative biomarkers of clinical outcomes on trastuzumab therapy, and that multiple subtypes of HER2-positive breast cancer may be defined by quantitative measurements of H3T, p95, and H2T.Electronic supplementary materialThe online version of this article (doi:10.1007/s10549-013-2665-0) contains supplementary material, which is available to authorized users.

Purpose: P95HER2 (p95) is a truncated form of the HER2, which lacks the trastuzumab-binding site and contains a hyperactive kinase domain. Previously, an optimal clinical cutoff of p95 expression for progression-free survival (PFS) and overall survival (OS) was defined using a quantitative VeraTag assay (Monogram Biosciences) in a training set of trastuzumab-treated metastatic breast cancer (MBC) patients.Experimental Design: In the current study, the predictive value of the p95 VeraTag assay cutoff established in the training set was retrospectively validated for PFS and OS in an independent series of 240 trastuzumab-treated MBC patients from multiple institutions.Results: In the subset of 190 tumors assessed as HER2-total (H2T)-positive using the quantitative HERmark assay (Monogram Biosciences), p95 VeraTag values above the predefined cutoff correlated with shorter PFS (HR ¼ 1.43; P ¼ 0.039) and shorter OS (HR ¼ 1.94; P ¼ 0.0055) where both outcomes were stratified by hormone receptor status and tumor grade. High p95 expression correlated with shorter PFS (HR ¼ 2.41; P ¼ 0.0003) and OS (HR ¼ 2.57; P ¼ 0.0025) in the hormone receptor-positive subgroup of patients (N ¼ 78), but not in the hormone receptor-negative group. In contrast with the quantitative p95 VeraTag measurements, p95 immunohistochemical expression using the same antibody was not significantly correlated with outcomes.Conclusions: The consistency in the p95 VeraTag cutoff across different cohorts of patients with MBC treated with trastuzumab justifies additional studies using blinded analyses in larger series of patients. Clin Cancer Res; 20(10); 2805-13. Ó2014 AACR.