Virtual filtering and screening of combinatorial libraries have recently gained attention as methods complementing the high-throughput screening and combinatorial chemistry. These chemoinformatic techniques rely heavily on quantitative structure-activity relationship (QSAR) analysis, a field with established methodology and successful history. In this review, we discuss the computational methods for building QSAR models. We start with outlining their usefulness in high-throughput screening and identifying the general scheme of a QSAR model. Following, we focus on the methodologies in constructing three main components of QSAR model, namely the methods for describing the molecular structure of compounds, for selection of informative descriptors and for activity prediction. We present both the well-established methods as well as techniques recently introduced into the QSAR domain.

Bronchiolitis obliterans (BrOb), a late complication of bone marrow transplantation (BMT), is associated with chronic graft-versus-host disease (GVHD) and is frequently fatal. To identify the risk factors associated with BrOb, the factors affecting survival, treatment outcomes, and causes of death of patients with BrOb, we retrospectively analyzed 2859 BMT recipients. No cases of BrOb occurred among 1070 autologous BMT recipients. Among 1789 allogeneic BMT recipients, we identified 47 patients with BrOb. In multivariate analysis, older recipients or donors and acute GVHD were significantly associated with the development of BrOb. Among patients with BrOb, 5-year survival from the time of transplantation was only 10%, versus 40% among allogeneic BMT recipients without BrOb. The clinical course of BrOb had a significant effect on survival: 79% survived 5 years from the time of BrOb diagnosis if BrOb improved versus 13% if there was no improvement after the first-line therapy. Predictors of response included older donors and recipients, a previous diagnosis of chronic GVHD, and diagnosis of BrOb 6 months after transplantation; each of these significantly increased the likelihood of a favorable response to treatment. BrOb had high mortality rate of 55%, and pulmonary failure was the leading cause of death. More effective BrOb therapy is needed, especially for patients with unfavorable presentation.

BACKGROUND: Sunitinib and sorafenib are small‐molecule tyrosine kinase inhibitors (TKI) with antitumor activity in advanced renal cell carcinoma. A retrospective study was conducted to assess the response of renal cell carcinoma to sequential treatment with these two agents.METHODS: Tumor response was evaluated by using Response Evaluation Criteria In Solid Tumors (RECIST) criteria in patients failing first‐line therapy with either sunitinib or sorafenib and subsequently receiving second‐line therapy with the other TKI agent.RESULTS: Twenty‐nine patients received sorafenib followed by sunitinib (Group A), and 20 patients received sunitinib followed by sorafenib (Group B). TKI drugs were terminated in 6 (12%) patients in Group A and 4 (8%) in Group B because of toxicity. Median duration of stable disease for Groups A and B was 20 and 9.5 weeks, respectively. Median time from starting first TKI to disease progression after second TKI (time to progression) in Groups A and B was 78 and 37 weeks, respectively. Multivariate analysis revealed that Group B had a shorter time to progression than Group A (risk ratio [RR] 3.0; P = .016). Median overall survival was 102 and 45 weeks in Groups A and B, respectively (P = .061).CONCLUSIONS: The longer duration of disease control in patients who received sorafenib followed by sunitinib warrants further investigation. Cancer 2009. © 2008 American Cancer Society.

BACKGROUND: Inhibiting src kinases (non‐receptor tyrosine kinase signaling intermediates) reduces melanoma cell proliferation and invasion. Dasatinib inhibits c‐kit, PDGFβR, and EPHA2 and src kinases c‐src, c‐Yes, Lck, and Fyn. A phase 2 trial of dasatinib in melanoma was conducted to assess response rate (RR), progression‐free survival (PFS), and toxicity.METHODS: Adults with stage 3/4 chemotherapy‐naïve unresectable melanoma were eligible. Dasatinib was initially administered at 100 mg twice daily continuously to 17 patients. Due to toxicity, the starting dosage was decreased to 70 mg twice daily. Tumor assessments occurred every 8 weeks.RESULTS: Thirty‐nine patients were enrolled, 36 of whom were evaluable for activity and toxicity. Five, 4, and 3 patients had acral‐lentiginous, ocular, or mucosal primaries, respectively. Two patients had confirmed partial responses lasting 64 and 24 weeks (RR 5%). Three patients had minor responses lasting 136, 64, and 28 weeks, and 1 patient who was responding discontinued due to noncompliance. The median PFS was 8 weeks; the 6‐month PFS rate was 13%. One patient with an exon‐13 c‐kit mutation had a partial response, whereas disease in another patient with an exon‐11 c‐kit mutation progressed. Common toxicities were fatigue, dyspnea, and pleural effusion.CONCLUSIONS: Daily dasatinib has minimal activity in unselected melanoma patients, excluding those with c‐kit mutations. The study did not meet the prespecified endpoints of 30% response rate or 6‐month PFS. Dasatinib was poorly tolerated overall, often requiring dose reduction or interruption. Because activity was observed in a small subset without c‐kit mutations, identifying predictive biomarkers is important for future development of dasatinib in melanoma alone or in combination trials. Cancer 2011. © 2010 American Cancer Society.

Platelet factor 4 (PF4) is an abundant platelet α-granule C-X-C chemokine that has weak chemotactic potency but strongly inhibits hematopoiesis through an unknown mechanism. We find that PF4 binds to human CD34+ hematopoietic progenitor cells (HPCs) with a median effective concentration of 1 μg/mL but not after exposure to chondroitinase ABC. PF4 enhances adhesion of HPCs to intact stroma. Committed progenitors also adhere avidly to immobilized PF4. This adhesion is time-dependent, requires metabolic activity, causes cytoskeletal rearrangement, and induces cell-cycle inhibition. Using extracellular acidification rate to indicate transmembrane signaling, we find that interleukin-8 (IL-8), but not PF4, activates CD34+ progenitors, and PF4 blocks IL-8–mediated activation. Surface plasmon resonance analysis shows that PF4 binds IL-8 with high (dissociation constant [Kd] = 42 nM) affinity. Nuclear magnetic resonance analysis of IL-8 and PF4 in solution confirms this interaction. We conclude that PF4 has the capacity to influence hematopoiesis through mechanisms not mediated by a classical high-affinity, 7-transmembrane domain chemokine receptor. Instead, PF4 may modulate the hematopoietic milieu both directly, by promoting progenitor adhesion and quiescence through interaction with an HPC chondroitin sulfate–containing moiety, and indirectly, by binding to or interfering with signaling caused by other, hematopoietically active chemokines, such as IL-8.