Citations

Abstract Abstract 800 Background: Ruxolitinib (RUX), an oral JAK1/JAK2 inhibitor, reduced spleen volume (SV), improved myelofibrosis (MF)-associated symptoms and quality of life (QoL), and appeared to exhibit a survival advantage over placebo (PBO) in patients (pts) with MF regardless of JAK2V617F mutation status in the phase III COMFORT-I study. We describe long-term efficacy and safety of RUX from COMFORT-I, with 1 year of additional follow up beyond previously published data. Methods: Eligible pts (N=309) were randomized (1:1) to RUX or PBO. The primary analysis occurred when all pts completed 24 weeks (wks) and when half the pts completed 36 wks of treatment. All pts receiving PBO were eligible for crossover to RUX after the primary analysis; crossover before wk 24 was permitted if pts met protocol-defined criteria for worsening splenomegaly. The proportion of pts with ≥35% SV reduction at 24 wks (primary endpoint) and durability of SV response were assessed. Although symptom burden (measured daily using the modified MF Symptom Assessment Form v2.0) was only measured up to wk 24, QoL continued to be evaluated beyond wk 24 (every 24 wks) using the EORTC QoL Questionnaire-Core 30 (QLQ-C30). Overall survival (OS) was assessed according to original randomized treatment. Results: In this updated analysis, median follow-up of pts randomized to RUX was 102 wks. All pts receiving PBO completed crossover or discontinued within 3 months of the primary analysis. Of 134 pts randomized to RUX who remained on treatment after the primary data analysis, 100 continue on study. Mean SV reduction in pts randomized to RUX was 31.6% at wk 24 and has remained stable with additional follow up through wk 96 (Table). In pts who achieved a ≥35% SV reduction, response was durable, with a median response duration of 108 wks. RUX treatment was also associated with durable improvements in the Global Health Status/QoL (Table) and the 5 functional domains of the EORTC QLQ-C30. Twenty-seven (27) pts randomized to RUX and 41 pts randomized to PBO died, representing a continued OS benefit in favor of RUX (HR=0.58; 95% CI: 0.36, 0.95; P = 0.028; Fig 1) similar in magnitude to that previously reported. OS favored RUX across subgroups including starting dose as well as baseline risk status and hemoglobin (Hgb). Of 34 pts randomized to RUX who discontinued after the primary analysis, 4 discontinued for an adverse event (AE). In pts who continued on RUX, anemia and thrombocytopenia remained the most frequently reported AEs. New onset of grade 3 or 4 anemia and thrombocytopenia was reported in only 12 and 5 pts, respectively. One pt discontinued for anemia. Overall, among all pts randomized to RUX, Grade 3 and 4 anemia regardless of baseline Hgb was reported in 37.4% and 14.8% of pts, respectively. Similarly, Grade 3 and 4 thrombocytopenia was reported in 11.0% and 5.2% of pts, respectively. These rates were similar to those reported in the primary analysis. By wk 36, the proportion of pts receiving red blood cell transfusions decreased to the level seen with PBO and remained stable thereafter (Fig 2). Rates of nonhematologic AEs adjusted for increased follow-up duration remain similar to those seen at the time of the primary data analysis. No additional cases of acute myeloid leukemia (AML) in pts randomized to RUX were reported. Two pts originally randomized to PBO developed AML, 21 and 178 days after crossover to RUX. There continued to be no reports of a withdrawal syndrome after RUX discontinuation. Conclusions: RUX provides durable reductions in SV and improvements in QoL. Although all pts randomized to PBO crossed over to RUX shortly after the primary analysis, with 1 year of additional follow up, RUX continues to be associated with a survival advantage over PBO. RUX continues to be well tolerated; the AE profile with long-term treatment is consistent with that previously reported. The proportion of pts receiving transfusions decreased over time to rates similar to PBO, and there were no reports of a specific withdrawal syndrome or cytokine rebound phenomenon after RUX discontinuation. Disclosures: Verstovsek: Incyte Corporation: Research Funding. Mesa:Incyte: Research Funding; Lilly: Research Funding; Sanofi: Research Funding; NS Pharma: Research Funding; YM Bioscience: Research Funding. Gotlib:Incyte: Consultancy, travel to congress Other. Levy:Incyte: Employment, Equity Ownership. Gupta:Incyte: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Research Funding; YM Biosciences: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Sanofiå]Aventis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Catalano:Incyte: Consultancy. Deininger:Bristol-Myers Squibb: Consultancy, Research Funding; Novartis: Consultancy; Ariad: Consultancy. Miller:Novartis: Consultancy, Research Funding, Speakers Bureau, development of educational presentations Other; Incyte: development of educational presentations, development of educational presentations Other. Talpaz:Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; B.M.S.: Membership on an entity's Board of Directors or advisory committees, Research Funding; Ariad: Research Funding; Sanofi: Membership on an entity's Board of Directors or advisory committees, Research Funding; Teva: Membership on an entity's Board of Directors or advisory committees; Pfizer: Membership on an entity's Board of Directors or advisory committees. Winton:Incyte: Consultancy, Honoraria. Arcasoy:Incyte: Research Funding. Lyons:Incyte: Consultancy, Research Funding; Amgen: Consultancy, Research Funding; Novartis: Research Funding; Telik: Research Funding. Paquette:Incyte: Consultancy. Vaddi:Incyte: Employment, Equity Ownership. Erickson-Viitanen:Incyte: Employment, Equity Ownership. Sun:Incyte Corporation: Employment, Equity Ownership. Sandor:Incyte Corporation: Employment, Equity Ownership. Kantarjian:Incyte: grant support Other.

Abstract Abstract 755 Background: TG101348 is a potent, orally bioavailable, JAK2-selective small molecule inhibitor, that is currently being evaluated in a Phase I study for the treatment of myelofibrosis. Data from the dose escalation cohort (n=28; 30-800mg administered as a single daily dose) showed dose-linear plasma exposure, with mean elimination T1/2 at steady state ranging from 20 to 52 hours. The dose-limiting toxicity was asymptomatic grade 3 or 4 amylasemia/lipasemia that was reversible, and the maximum tolerated dose (MTD) was 680mg. The most frequent non-hematological toxicities were mild nausea, vomiting, and/or diarrhea that were easily controlled or resolved spontaneously. Grade 3/4 neutropenia and thrombocytopenia were observed in 14% and 25% of patients, respectively. TG101348 had activity in reducing spleen size, leukocyte count, and JAK2V617F (VF) allele burden. Here, we present updated results with a focus on data from the dose confirmation cohort who initiated treatment at a dose of 680mg/day. Results: Fifty nine patients (median age=66 years; range 43-86) have been treated – 28 in the dose escalation phase, and 31 in the dose confirmation phase. Overall, 44 patients had PMF, 12 post-PV MF, and 3 post-ET MF; 86% were VF-positive. Median palpable spleen size was 18cm and 22 patients were RBC transfusion-requiring at study enrollment. After a median follow-up of 12 weeks (range <1-76), 18 (31%) patients have discontinued treatment due to toxicity (n=7; thrombocytopenia=3, neutropenia=1), comorbidities (n=5), withdrawal of consent (n=4), or non-compliance/lack of response (1 each). The remaining 41 patients are currently at the following dose levels: 680mg (n=14), 520-600mg (n=16), 360-440mg (n=10), and 240mg (n=1). The cumulative drug exposure to date is 362 patient-months; exposure at or above MTD (≥680mg) is 154 patient-months. Forty patients (68%) started treatment at ≥680mg. Toxicity: TG101348 is well tolerated. Of the patients who started at ≥680mg, Gr3/4 neutropenia was observed in 15/0% and Gr3/4 thrombocytopenia in 20/10%. Twenty four (60%) patients did not require RBC transfusions at baseline (median Hgb=9.6g/dL; range 7.4-13.1); of these, 42% and 8% of patients developed Gr3 and Gr4 anemia, respectively. The majority of patients who started at ≥680mg developed mild nausea (1 Gr3), vomiting (1 Gr3), and/or diarrhea (3 Gr3) that were self-limited or easily controlled. Other non-hematological toxicities included Gr1/2 transaminitis (38%), Gr1/2 serum creatinine elevation (38%), and asymptomatic hyperlipasemia (33%). Efficacy: Thirty three patients who started at ≥680mg have completed at least 3 cycles of treatment; at 3 months, reduction in palpable spleen size (baseline median=18cms; range 6-32) was at least 50% in 22 (67%) patients; the spleen became non-palpable in 9 (27%) patients. All 21 patients with leukocytosis at baseline (WBC range 11 to 203 ×109/L) who started at ≥680mg have experienced a marked reduction in their WBC count (range 4 to 90); 70% had a normal WBC count at their last follow-up visit. Overall, 48 of the 51 VF-positive patients completed at least 1 cycle and were evaluable for response in VF allele burden; at last available follow-up, the median decrease in granulocyte mutant allele burden was 48%; 21 (44%) patients have had a ≥50% reduction, and in the group who started treatment at ≥680mg, 48% have had a ≥50% reduction. Of those evaluable, there was clinically significant benefit or resolution of constitutional symptoms, including early satiety, fatigue, cough, pruritus, and night sweats. Conclusions: TG101348 continues to be well tolerated in patients with myelofibrosis. Spleen and leukocyte responses are frequent, observed early, and produce substantial clinical benefit for patients. These responses are associated with significant decrease in VF allele burden and point to activity of TG101348 against the malignant clone in myelofibrosis. Disclosures: Pardanani: TargeGen: Research Funding; Cytopia: Research Funding. Off Label Use: Data from ongoing clinical trial will be presented. Gotlib:TargeGen: Research Funding. Jamieson:Merck: Research Funding; Pfizer: Research Funding; Wintherix: Consultancy; TargeGen: Research Funding; Celgene: Research Funding. Cortes:Targegen: Research Funding. Stone:Cephalon: ad hoc advisory board. Silverman:TargeGen: Consultancy. Shorr:TargeGen: Employment, Equity Ownership. Gilliland:Merck: Employment. Tefferi:TargeGen: Research Funding.

Abstract Abstract 3992 Little is known about the rare entities of heavy and light chain amyloidosis (AHL) and heavy chain amyloidosis (AH). In this study, we report the renal and hematologic characteristics, pathology, and outcome of 17 patients with renal AH/AHL including 5 with AH (4 IgG and 1 IgA) and 12 with AHL (7 IgGλ, 3 IgAκ, 1 IgAλ, and 1 IgMλ), and compare them with 202 patients with renal AL amyloidosis (AL) diagnosed during the same time period. All cases were diagnosed by kidney biopsy that showed Congo red-positive deposits. Amyloid typing was done by laser microdissection and mass spectrometry (LMD/MS) (12 patients) or by immunofluorescence (5 patients). All patients with renal AH/AHL were Caucasians, with a M:F ratio of 2.4 and a median age at biopsy of 63 years. Compared with patients with renal AL, those with renal AH/AHL had less frequent concurrent cardiac involvement, higher likelihood of having circulating complete monoclonal Ig, lower sensitivity of fat pad biopsy and bone marrow biopsy for detecting amyloid, higher incidence of hematuria, and better patient survival. The hematologic and renal responses to chemotherapy were comparable to renal AL. In 42% of patients, AH/AHL could not have been diagnosed without LMD/MS. In conclusion, renal AH/AHL is an uncommon but under-recognized form of amyloidosis, and its diagnosis is greatly enhanced by the use of LMD/MS for amyloid typing. The accurate histological diagnosis of renal AH/AHL and distinction from AL may have important clinical and prognostic implications. Table 1. Demographics and hematologic characteristics AH/AHL AL p value No. of patients 17 202 Gender: Male/female 12/5 (71%/29%) 126/76 (62%/38%) 0.61 Age, median (range) 63 (50–77) 62 (36–86) 0.73 Additional organ involvement 8 (47%) 126 (62%) 0.3 Cardiac involvement 3 (18%) 100 (50%) 0.01* % of plasma cells in bone marrow, median (IQR) 8 (5–15) 6 (5–10) 0.82 ≥30 plasma cells 4 (24%) 11/198 (6%) 0.02* Positive SPEP/SIF for paraprotein 15 (88%) 158/200 (79%) 0.53 Presence of whole monoclonal protein on SPEP 14 (82%) 108/200 (54%) 0.04* Positive UPEP/UIF for paraprotein 13/16 (81%) 158/189 (84%) 0.73 Presence of whole monoclonal protein on UPEP 10/16 (63%) 61/189 (32%) 0.03* Abnormal serum FLC ratio (<0.26 or >1.65) 9/12 (75%) 150/188 (80%) 0.71 Markedly abnormal FLC ratio (< 0.125 or > 8) 5/12 (42%) 100/188 (53%) 0.55 Positive bone marrow for amyloid 6/16 (38%) 135/183 (74%) 0.004* Positive fat pad biopsy for amyloid 2/14 (14%) 105/145 (72%) <0.001* IQR, interquartile range. Table 2. Renal characteristics at kidney biopsy AH/AHL AL p value No. of patients 17 202 24h urine protein in g, median (IQR) 5.1 (3.2–9.0) 6.0 (3.2–10.0) 0.9 Full nephrotic syndrome 9/16 (56%) 132/197 (67%) 0.42 Serum albumin in g/dl, median (IQR) 2.7 (2.2–3.3) 2.5 (1.9–2.9) 0.29 % albuminuria on UPEP, median (IQR) 68 (61–72) 70 (60–76) 0.47 Serum creatinine in mg/dl, median (IQR) 1.4 (1.1–2.1) 1.2 (0.9–1.8) 0.25 Serum creatinine >1.2 mg/dl 10/16 (63%) 92/201 (46%) 0.3 eGFR, median (IQR) 47 (27–67) 58 (36–75) 0.29 Decreased eGFR 10/16 (63%) 103/201 (51%) 0.44 Microscopic hematuria 9/16 (56%) 44/169 (26%) 0.02* IQR, interquartile range. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 707 Background: Myeloproliferative neoplasms (MPNs) are disorders that result in unregulated overproduction of one or more myeloid blood cell types by the bone marrow. Polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF) comprise the three classic MPNs. The somatic JAK2 V617F mutation is present in 95% of PV patients, and 50–60% of ET and PMF patients. Past work has identified a germline haplotype in JAK2 associated with risk of developing a V617F-positive MPN (Kilpivaara et al; Olcaydu et al; Jones et al, Nat. Genet., 2009). Methods: We recruited a web-based participatory cohort of individuals with MPNs to better understand the genetic basis of these conditions. Participation in this MPN research initiative included free genotyping, and participants provided informed consent and responses to surveys online, under a protocol approved by the external AAHRPP-accredited IRB, Ethical & Independent Review Services. We enrolled and collected saliva samples from more than 700 participants with the following self-reported diagnoses: systemic mastocytosis (n=142), ET (n=145), PV (n=137), PMF (n=50), chronic myelogenous leukemia (n=108), and 97 with overlapping diagnoses (e.g. ET or PV with PMF). Participants were genotyped using a derivative of the Illumina HumanOmniExpress BeadChip with additional custom content, including probes for the JAK2 V617F variant. We performed a genome-wide association study (GWAS) of classic MPN diagnoses (at least one of ET, PV, or MF), using 407 cases from this MPN research initiative and 94,037 controls drawn from the broader 23andMe research participant community. Association was assessed by logistic regression adjusted for age, gender, and 5 principal components to model ancestry. Results: We replicated the known association between germline variation in JAK2 and classic MPNs (rs12340895: odds ratio=2.5, P=7.4 × 10−35). We also identified a novel genome-wide-significant association in TERT, or telomerase reverse transcriptase (rs2853677: odds ratio=1.6 per G allele, P=6.3 × 10−11). The association is consistent with an additive model on the logistic scale (odds ratio=2.6 for 2 versus 0 G alleles). Another variant in TERT, rs2736100, has previously been associated with several non-hematological cancers and with red blood cell count, and our lead SNP is in linkage disequilibrium with TERT variant rs2736100 (r2=0.54). While no other loci reached the threshold of genome-wide significance (P<5.0 × 10−8), we saw a suggestive association with a germline missense variant (F858L) in ATM, or ataxia telangiectasia mutated gene (rs1800056: odds ratio=2.6 per C allele, P=1.4 × 10−5). This variant has also been reported to be associated with breast cancer and chronic lymphocytic leukemia. The rs2853677 TERT variant was associated with each of ET, PV, and PMF, with similar effect sizes. It was not associated with JAK2 V617F mutation status within the MPN cohort (odds ratio=1.0, P=1.0), while the germline JAK2 variant was preferentially associated with V617F-positive MPNs (odds ratio=2.1, P=3.0 × 10−6). The high-risk allele of rs2853677 was also predictive of V617F status among 23andMe participants who had not been recruited into the MPN community (odds ratio=1.6, P=0.0012). These results suggest that while variation in TERT facilitates either acquisition or proliferation of V617F-positive myeloid cells, it also predisposes to MPNs through a V617F-independent mechanism. Conclusion: We have identified a germline variant in TERT which is a new predisposition allele for classic MPNs. These results demonstrate the potential for web-based recruitment methods to contribute to genetic research for uncommon diseases. Disclosures: Hinds: 23andMe: Employment, Equity Ownership, Patents & Royalties. Barnholt:23andMe, Inc.: Employment. Zehnder:23andMe, Inc.: Unpaid advisor and collaborator Other. Kiefer:23andMe, Inc.: Employment. Do:23andMe, Inc.: Employment, Equity Ownership, Patents & Royalties. Eriksson:23andMe, Inc.: Employment, Equity Ownership, Patents & Royalties. Mountain:23andMe, Inc.: Consultancy, Employment, Honoraria, Patents & Royalties, Research Funding. Francke:23andMe, Inc.: Employment, Honoraria, Research Funding; Locus Development: Consultancy, Membership on an entity's Board of Directors or advisory committees. Tung:23andMe, Inc.: Employment. Mesa:23andMe, Inc: Unpaid advisor and collaborator Other. Gotlib:23andMe, Inc.: Unpaid advisor and collaborator Other.

INTRODUCTION: Volumetric laser endomicroscopy (VLE) is a high-resolution imaging modality used in Barrett’s esophagus (BE) surveillance. Established VLE scoring systems use a combination of features to guide BE dysplasia diagnosis but it remains unclear whether these features are sufficient to reach optimal VLE diagnostic performance. This study explores an exhaustive list of VLE features using one-to-one imaging to pathology correlation to develop optimal VLE diagnostic models for BE dysplasia. METHODS: In this prospective, multi-center (5 sites) study, patients with dysplastic BE underwent VLE with laser targeted regions of interest (ROIs), followed by standard surveillance endoscopy and biopsy directed to laser-marked sites. Histology was reviewed by two GI pathologists (MV, RO) who achieved a consensus diagnosis (gold-standard). ROIs were dysplastic if low-grade dysplasia (LGD), high-grade dysplasia (HGD) or intramucosal adenocarcinoma (IMC) was present. Two VLE experts (CLL, GJT) reviewed ROIs and corresponding endoscopic images to identify and define a set of VLE features associated with BE dysplasia. A multi-variable logistic regression model based on least absolute shrinkage and selection operator (LASSO) method was developed to determine the optimal diagnostic performance of these features. A decision tree model with high clinical usability also was developed to evaluate VLE diagnostic performance for real-time image interpretation. RESULTS: A total of 461 VLE ROIs with matching histology were obtained from 150 patients (Table 1). A final dataset of 55 neoplastic (LGD, 12; HGD, 24; IMC, 19) and 56 randomly selected non-dysplastic ROIs were reviewed. Twenty-eight VLE features were identified describing layering, signal intensity, surface topography, and glandular characteristics. LASSO logistic regression modeling identified 5 VLE features which when utilized optimized diagnostic performance with an area under the curve (AUC) for BE dysplasia of 0.95 (Figure 1). To improve clinical usability, a decision tree model was optimized to three levels with an AUC of 0.90 (Figure 2). CONCLUSION: This study represents the most comprehensive review of VLE features associated with BE dysplasia to date. The proposed high-performing diagnostic models suggest that VLE can be used efficiently and effectively to enhance BE surveillance by identifying ROIs for targeted biopsy. Further validation of the proposed VLE models will determine their clinical utility.

The NCCN Guidelines for Multiple Myeloma provide recommendations for diagnosis, workup, treatment, follow-up, and supportive care for patients with monoclonal gammopathy of renal significance, solitary plasmacytoma, smoldering myeloma, and multiple myeloma. These NCCN Guidelines Insights highlight some of the important updates and changes in the 1.2020 version of the NCCN Guidelines for Multiple Myeloma.

Aims A number of methods have been described to remove a well-fixed humeral implant as part of revision shoulder arthroplasty. These include the use of cortical windows and humeral osteotomies. The router bit extraction technique uses a high-speed router bit to disrupt the bone-implant interface. The implant is then struck in a retrograde fashion with a square-tip impactor and mallet. The purpose of this study was to determine the characteristics and frequency of the different techniques needed for the removal of a well-fixed humeral stem in revision shoulder arthroplasty. Patients and Methods Between 2010 and 2018, 288 revision shoulder arthroplasty procedures requiring removal of a well-fixed humeral component were carried out at a tertiary referral centre by a single surgeon. The patient demographics, indications for surgery, and method of extraction were collected. Results Of the 288 revisions, 284 humeral stems (98.6%) were removed using the router bit extraction technique alone. Four humeral stems (1.39%) required an additional cortical window. Humeral osteotomy was not necessary in any procedure. Most of the humeral stems removed (78.8%) were cementless. Of the four humeral stems that required a cortical window, three involved removal of a hemiarthroplasty. Two were cemented and two were cementless. Conclusion The router bit extraction technique removed a well-fixed humeral component in a very high proportion of patients (98.6%). This method allows surgeons to avoid more invasive approaches involving a cortical window or humeral osteotomy, and their associated complications. Cite this article: Bone Joint J 2019;101-B:1280–1284

Abstract Purpose: The surface receptor MET is highly expressed on primary uveal melanoma; MET inhibitors demonstrated early clinical signals of efficacy in slowing uveal melanoma growth. The primary objective of our study was to compare the progression-free survival rate at 4 months (PFS4) of patients with uveal melanoma treated with cabozantinib or chemotherapy. Patients and Methods: Patients with metastatic uveal melanoma and RECIST measurable disease were randomized 2:1 to receive either cabozantinib (arm 1) versus temozolomide or dacarbazine (arm 2) with restaging imaging every two cycles. Cross-over from arm 2 to cabozantinib after progression was allowed (arm 2X). Available tumor specimens were analyzed by whole-exome sequencing (WES) and results were correlated with outcome. Results: Forty-six eligible patients were accrued with 31, 15, and 9 in arms 1, 2, and 2X, respectively. Median lines of prior therapy, including hepatic embolization, were two. Rates of PFS4 in arm 1 and arm 2 were 32.3% and 26.7% (P = 0.35), respectively, with median PFS time of 60 and 59 days (P = 0.964; HR = 0.99). Median overall survival (OS) was 6.4 months and 7.3 months (P = 0.580; HR = 1.21), respectively. Grade 3–4 Common Terminology Criteria for Adverse Events were present in 61.3%, 46.7%, and 37.5% in arms 1, 2, and 2X, respectively. WES demonstrated a mean tumor mutational burden of 1.53 mutations/Mb and did not separate OS ≤ or &gt;1 year (P = 0.14). Known mutations were identified by WES and novel mutations were nominated. Conclusions: MET/VEGFR blockade with cabozantinib demonstrated no improvement in PFS but an increase in toxicity relative to temozolomide/dacarbazine in metastatic uveal melanoma.

Abstract Purpose: We hypothesized that integrated analysis of cancer types from different lineages would reveal novel molecularly defined subgroups with unique therapeutic vulnerabilities. On the basis of the molecular similarities between subgroups of breast and ovarian cancers, we analyzed these cancers as a single cohort to test our hypothesis. Experimental Design: Identification of transcriptional subgroups of cancers and drug sensitivity analyses were performed using mined data. Cell line sensitivity to Hsp90 inhibitors (Hsp90i) was tested in vitro. The ability of a transcriptional signature to predict Hsp90i sensitivity was validated using cell lines, and cell line- and patient-derived xenograft (PDX) models. Mechanisms of Hsp90i sensitivity were uncovered using immunoblot and RNAi. Results: Transcriptomic analyses of breast and ovarian cancer cell lines uncovered two mixed subgroups comprised primarily of triple-negative breast and multiple ovarian cancer subtypes. Drug sensitivity analyses revealed that cells of one mixed subgroup are significantly more sensitive to Hsp90i compared with cells from all other cancer lineages evaluated. A gene expression classifier was generated that predicted Hsp90i sensitivity in vitro, and in cell line- and PDXs. Cells from the Hsp90i-sensitive subgroup underwent apoptosis mediated by Hsp90i-induced upregulation of the proapoptotic proteins Bim and PUMA. Conclusions: Our findings identify Hsp90i as a potential therapeutic strategy for a transcriptionally defined subgroup of ovarian and breast cancers. This study demonstrates that gene expression profiles may be useful to identify therapeutic vulnerabilities in tumor types with limited targetable genetic alterations, and to identify molecularly definable cancer subgroups that transcend lineage.