McCullough, Ann E.

Abstract Background: Breast tumors exhibit intratumor heterogeneity resulting in targeted therapy resistance and other challenges in disease management. To address the sources of heterogeneity, we performed a unique, in-depth analysis of clonal architecture in primary chemoradiation-naïve breast cancers. We combined DNA content-based flow cytometry and ploidy analysis with aCGH and whole exome next-generation sequencing (NGS) in multiple biopsies from the tumors and involved lymph nodes (LNs). Material and methods: We sorted nuclei from distinct populations of diploid, tetraploid, and aneuploid cells in surgical tumor samples from three chemoradiation-naïve patients. Each sorted tumor cell population was interrogated with aCGH and whole exome NGS. In Patient #1, we sorted and interrogated the genomes of tumor populations from 12 fresh frozen sections morphologically mapped from within a HER2+, ER+, PR- primary invasive ductal carcinoma (IDC) of histological grade 3 with LN involvement and 2-3 sections from 2 out of 5 LNs. In Patient #2, 11 morphologically mapped fresh frozen sections were analyzed from a grade 2, ER+, PR+, HER2+, BRCA2 mutant LN- IDC. In parallel, matching samples were processed for IHC assays. In Patient #3 biopsies from a grade 2 ER+, PR+ primary tumor were profiled. Results: The 18 primary and LN biopsies from Patient #1 fell into 6 distinct ploidy groups albeit with aberrant but homogenous aCGH profiles, characterized by SARC amplification and homozygous deletions in ROBO1 and ROBO2. In contrast a dominant ploidy was identified throughout 10 biopsies in Patient #2 but with heterogeneous aCGH profiles. Patient #3 had a clonal homozygous deletion in Numb in each of 4 tumor biopsies. Mutation profiles obtained through exome sequencing further confirmed that ploidy was the main driver in Patient #1 whereas copy number aberrations played the key role in Patient #2 with the BRCA2 mutation (R3129X). A dendrogram based on exome variant calls of the aneuploid populations in Patient #1 strongly correlated with ploidy group and further revealed the specific clonal population characterized by a 5N ploidy and homozygous mutations in TP53 and PIK3CA as the progenitor to the ploidies present in the distant LNs. Strikingly, patients #1 and #2 were HER2 wild type across their clonal populations, contradicting IHC staining in a single core biopsy. Conclusions:Rather than inferring the presence of distinct tumor cell populations, our flow-sorting based approach of first identifying clonal populations and then interrogating their genomes, provides an objective method of exploring the sources and clinical significance of tumor heterogeneity. Our approach of clonal analysis has broad implications in the study of tumor heterogeneity and the identification of drivers in breast and other solid tumors that can advance more effective treatment and clinical management of patients with this disease. Citation Format: MARIACARLA ANDREOZZI, Princy Francis, Elizabeth Lenkiewicz, Mia Champion, Brady Laughlin, Karen Anderson, Heather Cunliffe, Ann E. McCullough, Michael T. Barrett, Barbara Pockaj. Deep clonal profiling identifies distinct mechanisms of heterogeneity and evolution in breast cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr LB-175. doi:10.1158/1538-7445.AM2015-LB-175

Abstract Background: Genomic assessment of cancer has been revolutionized by Next-Generation sequencing and is increasingly being applied to guide clinical prognostic and therapeutic decision-making. Initial clinical applications have been limited to gene panels and whole exome based strategies due to challenges with time to reporting of results, specimen quantity, analyte quality, and ethical-legal-social issues (ELSI). Methods: Excisional or core tumor biopsies or bone marrow biopsies were obtained from consenting participants. Nucleic acid was extracted from fresh or frozen samples followed by WGS/WES/RNASeq on the Illumina HiSeq2000 or HiSeq2500 and bioinformatics analysis. Therapeutic targets were then prioritized by a multi-disciplinary Genomic Tumor Board (GTB) and CLIA validated using Sanger sequencing, RT-qPCR, FISH and/or IHC. Treatment was delivered using on/off-label FDA approved drugs, available clinical trials and single patient INDs. Results: We consented 64 and enrolled 35 patients with advanced, treatment-refractory cancers. Median age was 59 (range 27-91) with 62% male. The majority had ECOG scores of 1 (94%). A median of 79 (range 28-8891) potentially functional somatic point mutations were identified in each case. One to two mutations/case were further identified as therapeutically targetable in 60% of the cases. The median time from tissue acquisition to CLIA validated results was 116 days (range 42-282) with CLIA validation of targets achieved in 21 of 22 patients. Genomic, target directed treatment was ultimately instituted in 13 patients utilizing: on/off label FDA approved drugs (n = 9), clinical trial (n = 3) and single patient IND (n = 1). Preliminary clinical efficacy was noted in 5 patients (2 PR, 3 SD). Integration of WES, long-insert WGS and RNA-Seq identified a case with an ERRFI1 mutant allele present in only 11% of the DNA reads while 82% of the RNA-Seq reads had the mutant transcript. The patient was treated with erlotinib and achieved a partial response by RECIST criteria. In another case, a fusion between FGFR2-MGEA5 was observed in WES and RNA-Seq data leading to prioritization as a drug target. Reasons for patients not receiving targeted therapy included: inability to access treatment (n = 1), death prior to intended treatment (n = 3), results returned after death (n = 3), no targets identified (n = 2) and decision to pursue alternate therapy (n = 5). Conclusions: Integrating whole genome analysis in a CLIA setting is not only feasible, but also valuable, in the prioritization and selection of potential targeted therapies for patients with advanced tumors. Continued barriers to broad application include the need for shorter time to reporting as well as broad availability of therapies through basket studies. Citation Format: Mitesh J. Borad, Jan Egan, Mia Champion, Katherine Hunt, Robert McWilliams, Ann McCullough, Jessica Aldrich, Sara Nasser, Winnie Liang, Michael Barrett, David Craig, Ramesh Ramanathan, John Carpten, A. Keith Stewart, Alan Bryce. Implementation of CLIA enabled integrated whole genome (WGS)/exome (WES)/transcriptome (RNAseq) next-gen sequencing to identify therapeutically relevant targets in advanced cancer patients. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr CT112. doi:10.1158/1538-7445.AM2015-CT112

Abstract Introduction: Overexpression and amplification of PD-L1 has been reported in a subset of cancers, including lymphoma, and gastric cancer. We have recently described amplification of chromosome 9p24 targeting PD-L1 in a subset of triple negative breast cancers (TNBC), as well as 1-3% of colorectal carcinomas and glioblastomas. The clinical impact of these amplifications is not known. Methods: Tumor samples from 41 subjects with TNBC were evaluated for copy number aberrations. We performed DNA-based flow cytometry to sort nuclei from diploid, tetraploid, and aneuploid populations of tumor cells. The genomes of each sorted tumor cell population were then interrogated with oligonucleotide CGH arrays. Copy number intervals were identified and their genomic boundaries mapped using a step gram algorithm. Clinical data was available on 36 of 45 (80%) patients. Results: A recurring top ranked high level (log2ratio >1) amplicon that spanned 9p24.1 was detected in 12/41 (29.3%) of TNBCs. In contrast this amplicon was absent in ER+ (n = 8) and Her2+ (N = 15) breast tumors. The shortest region of overlap (SRO) spanned 777 kb and always included the PD-1 ligands PD-L1, PD-L2, and Janus kinase 2 (JAK2) loci, termed the “PDJ” amplicon. Additional genes in this amplicon include RCL1, MIR101-2, INSL6, INSL4, RLN2, RLN1, and C9orf46. The height of the PDJ amplicon included mean log2ratios of >4, consistent with genomic drivers such as Her2 and Myc. Patients with PD-L1 amplification (n = 8) were noted to have larger tumors (mean 3.9 vs. 1.9 cm, p = 0.04) and a higher incidence of lymph node metastases (75% vs. 26%, p = 0.01). Lymphocytic infiltration was only noted in 4 patients, all without PDJ amplification. Almost all patients received chemotherapy after definitive surgical therapy. The recurrence rate at 5 years was 62.5% in the PDJ amplified patients, and 25% in the unamplified patients (p = 0.02). Overall survival at 5 years was 37.5% in the PDJ amplified patients, compared with 78.2% in the unamplified patients (p = 0.02). Conclusion: We identified a focal amplification of chromosome 9p24.1 involving PD-L1, PD-L2, and JAK2 in a significant proportion of TNBCs. The high level of amplification and inverse correlation with survival suggests that the PDJ amplicon is a genetic driver event in these cancers, and warrants confirmation in larger studies. Citation Format: Karen S. Anderson, Christine L. Klassen, Amylou C. Dueck, Ann E. McCullough, Donald W. Northfelt, Mariacarla Andreozzi, Barbara A. Pockaj, Michael T. Barrett. High-level amplification of chromosome 9p24 targeting PD-L1 and JAK2 correlates with worse DFS and OS in triple negative breast cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2861. doi:10.1158/1538-7445.AM2015-2861

Abstract Background: While the biological functions and clinical importance of ERα are well understood in breast cancer, much less is known about its most closely related family member, ERβ, particularly in the setting of triple negative disease. Additionally, the ability to therapeutically target ERβ in triple negative breast cancer (TNBC) has not been fully explored. Methods: Expression of ERβ protein was determined using a well characterized and validated ERβ specific monoclonal antibody that only recognizes the full-length form of this receptor (PPG5/10) in a cohort of 71 TNBCs. To further define the biological functions of ERβ in TNBC, novel ERβ expressing triple negative cell lines (MDA-MB-231 and Hs578T) were developed and comprehensively characterized at the level of global gene expression profiling, modulation of important biological pathways, cellular proliferation and response to targeted therapies. Results: In TNBCs from 71 patients, nuclear and cytoplasmic ERβ was detected at moderate to high levels in 24% and 32% of cases respectively. This moderate to high expression of both nuclear and cytoplasmic ERβ was associated with higher levels of Ki67. Of the 17 tumors expressing ERβ, 13 (76%) were negative for androgen receptor expression. In the triple negative MDA-MB-231-ERβ and Hs578T-ERβ cell lines, expression of ERβ led to inhibition of proliferation in response to both estrogen and multiple ERβ specific agonists. Microarray analysis and RT-PCR profiling of these cells revealed that estrogen and ERβ agonists highly induced the expression of multiple cystatins, a family of small secreted cysteine protease inhibitors, while suppressing the expression of many interleukins. Conditioned media isolated from estrogen or ERβ agonist treated MDA-MB-231-ERβ cells inhibited the proliferation rates and blocked TGFβ signaling in non-ERβ expressing TNBC cells, effects that were completely reversed following depletion of cystatins from the conditioned media. Conclusions: ERβ is expressed in a substantial proportion of TNBCs, and most do not express the androgen receptor. In TNBC, where targeted therapies are lacking, our data suggest that estrogen or ERβ specific agonists would be expected to elicit anti-tumor effects when ERβ is expressed. These tumor suppressive effects of ERβ appear to be mediated in part through the actions of cystatins, and their inhibition of TGFβ signaling, suggesting that this family of secreted proteins may represent novel biomarkers for monitoring ERβ specific drug responsiveness and/or patient outcomes. Citation Format: Jordan M Reese, Malayannan Subramaniam, Vera J Suman, Xianglin Wu, Vivian Negron, Anne Gingery, Kevin S Pitel, Sejal S Shah, Heather E Cunliffe, Ann E McCullough, Barbara A Pockaj, Thomas C Spelsberg, Matthew P Goetz, James N Ingle, John R Hawse. Therapeutic targeting of ERβ in triple negative breast cancer [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P6-03-06.

Abstract Background: The development of novel chemotherapeutic agents has significantly improved the prognosis and survival of patients with breast cancer. ER positive or luminal tumors represent around two thirds of all breast cancers and these cancers are comprised of different histologies including differing gene expression and mutational profiles. This study examined biomarkers involved in immune evasion including PD-L1 and its association with other biological pathways as potential treatment options for luminal breast cancer patients. Methods: We analyzed 1311 breast samples using a multiplatform approach including whole genome mRNA expression (HumanHT-12 v4 BeadChip Illumina Inc., San Diego, CA), protein expression (immunohistochemistry), gene copy number changes (in situ hybridization) and gene sequencing and an additional 304 breast samples were tested for PD-1 and PD-L1 by IHC. The mRNA expression data was based on whole tumor and represents cell type heterogeneity. Heat map analysis was done to look at differential gene expression between the PD-L1 high vs low luminal breast cancers. Results: Based on expression of ER, PR and HER2 by IHC, we subdivided the data into sub- cohorts. Elevated mRNA expression of immune markers including PD-L1, CTLA4, B7H-3, and IDO1 was noted in the ER+HER2- luminal population including ER+ PR- and ER+PR+ cohort. Positive correlation was found between PD-L1 and other immune regulators including CTLA-4, B7-H3 and IDO1 (Spearman correlations of 0.49, 0.36 and 0.46). Protein expression of PD-L1 was found to be specific to the HER2 negative cohort with no expression in the HER2 positive cohort regardless of ER status. Within the ER+HER2- cohort, PD-L1 expression was 5% {ER+PR+ was 5% (4/81) and the ER+PR- was 6% (2/34)}. In contrast, PD-L1 expression was higher in the triple negative cohort, 17% (13/75). The expression of PD-1 on the other hand was present throughout the different cohorts. PD-1 expression ranged from 43% in the ER+HER2- cohort {41% (14/34) in ER+PR-HER2-; 43% (35/81) in ER+PR+HER2-} to 33% in the ER+HER2+ cohort {44% (4/9) in ER+PR-Her2+; and 17% (1/6) in ER+PR+HER2+}. In the ER-HER2+ cohort PD-L1 expression was 75%{6/8 in ER-PR-HER2+ and 0/0 in ER-PR+HER+} and 63%(47/75) in the triple negative cohort. Pathway analysis of the PD-L1 negative vs positive luminal population identified 127 genes involved in various cancer pathways including EGFR and VEGF signaling network (P=7.47e-09). Conclusions: The expression of immune regulatory targets in the breast cancer population suggests that immune- targeted therapies with anti PD-1/PD-L1, CTLA4, B7-H3 and IDO1 may be effective especially in the luminal cohort. Our study further shows that ER+HER2- breast cancer patients may be better candidates for immunotherapy with checkpoint inhibitors as compared to ER+HER2+ due to lack of PD-L1 expression in the HER2 positive cohort. Further validation of our findings is ongoing. Citation Format: Gargi D Basu, Anatole Ghazalpour, Randal Vader, Sandeep Reddy, Karen Anderson, Ann McCullough, Barbara Pockaj. Expression of novel immunotherapeutic targets in luminal breast cancer patients [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P5-04-08.

Abstract Background: Breast tumors exhibit intratumor heterogeneity resulting in targeted therapy resistance and other challenges in disease management. To address the sources of heterogeneity, we performed a unique, in-depth analysis of clonal architecture in primary chemoradiation-naïve breast cancers. We combined DNA content-based flow cytometry and ploidy analysis with aCGH and next-generation sequencing (NGS) in multiple biopsies from the tumors and involved lymph nodes (LNs). Material and methods: We used DNA content-based flow sorting to isolate nuclei from distinct populations of diploid and aneuploid tumor cells in surgical tumor samples from two chemoradiation-naïve patients. Each sorted tumor cell population was interrogated with aCGH and exome NGS. In Patient #1, we used 12 fresh frozen sections morphologically mapped from within a HER2+, ER+, PR- primary invasive ductal carcinoma (IDC) of histological grade 3 with LN involvement and 2-3 sections from 2 out of 5 LNs. In Patient #2, 11 morphologically mapped fresh frozen sections were analyzed from a grade 2, ER+, PR+, HER2+, BRCA2 mutant LN- IDC. In parallel, matching samples were processed for IHC assays. Results: We identified multiple co-existing aneuploid populations within the biopsies. The 17 primary and LN biopsies from Patient #1 fell into 6 distinct ploidy groups albeit with aberrant but homogenous aCGH profiles, characterized by SARC amplification and homozygous deletions in ROBO1 and ROBO2. In contrast a dominant ploidy was identified throughout Patient #2 but with heterogeneous aCGH profiles. Mutation profiles obtained through exome sequencing further confirmed that ploidy was the main driver in Patient #1 whereas copy number aberrations played the key role in Patient #2 with the BRCA2 mutation (R3129X). A dendrogram based on exome variant calls of the aneuploid populations in Patient #1 strongly correlated with ploidy group and further revealed the specific clonal population characterized by a 5N ploidy and homozygous mutations in TP53 and PIK3CA as the progenitor to the ploidies present in the distant LNs. Strikngly, both patients had no HER2 amplification or mutation across their clonal populations, contradicting the initial IHC staining in a single core biopsy. Conclusions Rather than inferring the presence of distinct tumor cell populations, our novel flow-sorting based approach of first identifying the clonal populations and then interrogating their genomes, provides an objective method of exploring the sources and clinical significance of tumor heterogeneity. Our approach of clonal analysis has broad implications in the study of tumor heterogeneity and the identification of drivers in breast and other solid tumors that can advance more effective treatment and clinical management of patients with this disease. Citation Format: Princy Francis, Lucretia M Alvarez, Elizabeth Lenkiewicz, Mia Champion, Lisa Evers, Karen L Anderson, Ann E McCullough, Michael T Barrett, Barbara Pockaj. Deep clonal profiling identifies distinct mechanisms of heterogeneity and evolution in breast cancer [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P2-03-07.

Abstract Background: Whole genome/exome/RNA sequencing has revolutionized the ability to assess the genomic landscape of cancer and is increasingly being utilized for clinical decision-making. Initial clinical applications have been constrained by specimen quantity, analyte quality and the time from sample acquisition to results report. Methods: Patients with advanced cancers underwent surgical resection, excisional/core biopsies, or bone marrow biopsy. Samples were analyzed by whole genome or exome sequencing in addition to RNA sequencing, bioinformatics analysis, and therapeutic target prioritization by a multi-disciplinary Clinical Genomics Board. All prioritized targets were CLIA validated using Sanger sequencing, RT-qPCR, FISH, or IHC as appropriate. Treatment was delivered using off-label FDA approved drugs, clinical trials, or single patient INDs. Results: We have enrolled 40 patients for whom sequencing data is available on 33. The initial 6 patients were evaluated in a non-CLIA pilot phase and 27 in a CLIA-enabled phase. Tumor types in the CLIA-enabled phase with the highest enrollment were pancreatic cancer (n=8) and cholangiocarcinoma (n=8). We sought to quantify the targets identified along with clinical benefit, defining these as the “Actionable Index” (AI) (proportion of patients with ≥ 1 putative drug target) and “Utility Index” (UI) (proportion of patients who derive clinical benefit). Putative therapeutic targets were identified in 7/8 (AI=0.88) cholangiocarcinoma (CC) patients and in 5/8 (AI=0.63) pancreatic cancer (PC) patients. All 3 CC patients who received target directed treatment achieved a partial response (UI=0.38). In contrast, none of the 4 PC patients who received target directed therapy had treatment response (UI=0.0). Interestingly no actionable targets were identified in 1 CC and in 2 PCs. One CC with an identified target was unable to access the drug and subsequently died. A CC patient and a PC patient, each with identified targets, expired prior to the initiation of therapy. Conclusions: While whole genome/exome/RNA sequencing is providing unparalleled detail of tumor genomes, the application to the clinic must be carefully considered. Actionability of targets will eventually need to be defined in close relation to eventual clinical utility and appropriate refinements to disease-gene-drug databases implemented. Preliminary observations in pancreatic cancer and cholangiocarcinoma demonstrate disparity in correlation between utility indices and actionable indices. Application of these tools in larger cohorts and types of tumors will need to be conducted to ascertain more precise estimates. Additional measures that are organ-site agnostic but pertain to specific targets (e.g. BRAF) will also need to be developed in order to facilitate more judicious application of sequencing in the clinical setting. Citation Format: Jan B. Egan, Alan H. Bryce, Mia D. Champion, Winnie S. Liang, Rafael Fonseca, Ann E. McCullough, Michael T. Barrett, Katherine Hunt, Rachel M. Condjella, Robert R. McWilliams, Stephen D. Mastrian, Janine LoBello, Daniel Von Hoff, David W. Craig, A. Keith Stewart, John D. Carpten, Mitesh J. Borad. Indices of actionability and clinical utility in a CLIA-enabled study of whole genome/exome/RNA sequencing in 33 cancer patients: Actionable vs. utility. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4694. doi:10.1158/1538-7445.AM2014-4694

Purpose To develop a genomic signature that predicts benefit from trastuzumab in human epidermal growth factor receptor 2–positive breast cancer. Patients and Methods DASL technology was used to quantify mRNA in samples from 1,282 patients enrolled onto the Combination Chemotherapy With or Without Trastuzumab in Treating Women With Breast Cancer (North Central Cancer Treatment Group N9831 [NCCTG-N9831]) adjuvant trastuzumab trial. Cox proportional hazard ratios (HRs), adjusted for significant clinicopathologic risk factors, were used to determine the association of each gene with relapse-free survival (RFS) for 433 patients who received chemotherapy alone (arm A) and 849 patients who received chemotherapy plus trastuzumab (arms B and C). Network and pathway analyses were used to identify key biologic processes linked to RFS. The signature was built by using a voting scheme. Results Network and functional ontology analyses suggested that increased RFS was linked to a subset of immune function genes. A voting scheme model was used to define immune gene enrichment based on the expression of any nine or more of 14 immune function genes at or above the 0.40 quantile for the population. This model was used to identify immune gene–enriched tumors in arm A and arms B and C. Immune gene enrichment was linked to increased RFS in arms B and C (HR, 0.35; 95% CI, 0.22 to 0.55; P < .001), whereas arm B and C patients who did not exhibit immune gene enrichment did not benefit from trastuzumab (HR, 0.89; 95% CI, 0.62 to 1.28; P = .53). Enriched immune function gene expression as defined by our predictive signature was not associated with increased RFS in arm A (HR, 0.90; 95% CI, 0.60 to 1.37; P = .64). Conclusion Increased expression of a subset of immune function genes may provide a means of predicting benefit from adjuvant trastuzumab.

We report a case of sinonasal phosphaturic mesenchymal tumor (PMT) and conduct a systematic review of the literature to highlight a unique paraneoplastic syndrome associated with PMT. We used English language publications from Medline and Cochrane databases (1970–2013) as data sources. A systematic review of the literature was conducted. All reported cases of head and neck PMTs were included. The presence or absence of the associated paraneoplastic syndrome was noted. We found 33 cases of PMT in the head and neck reported in the literature, 17 of which occurred in the sinonasal area. Approximately 5% of all PMTs are located in the head and neck. Just greater than half are concentrated in the sinonasal area, and the remaining involve various bony and soft tissue structures of the head and neck. PMT is sometimes associated with a paraneoplastic syndrome of tumor-induced (oncogenic) osteomalacia (TIO) causing bone pain, muscle weakness, and pathologic fractures. We present the 18th reported case of sinonasal PMT. A smooth mucosa-covered midline intraseptal mass filling the posterior nasal cavity with destruction and erosion of the skull base was found in an adult male. The patient underwent successful endoscopic resection with wide negative margins and is without recurrence at 24-month follow-up. PMT is a benign, locally aggressive tumor with rare malignant transformation. Knowledge of the bony invasion and destruction caused by this tumor is essential in planning surgical resection with wide negative margins. Familiarity with the associated TIO is essential to investigate for and manage any associated bony morbidity.

SummaryAs invasive mucormycosis (IM) numbers rise, clinicians suspect prior voriconazole worsens IM incidence and severity, and believe combination anti‐fungal therapy improves IM survival. To compare the cumulative incidence (CI), severity and mortality of IM in eras immediately before and after the commercial availability of voriconazole all IM cases from 1995 to 2011 were analysed across four risk‐groups (hematologic/oncologic malignancy (H/O), stem cell transplantation (SCT), solid organ transplantation (SOT) and other), and two eras, E1 (1995–2003) and E2, (2004–2011). Of 101 IM cases, (79 proven, 22 probable): 30 were in E1 (3.3/year) and 71 in E2 (8.9/year). Between eras, the proportion with H/O or SCT rose from 47% to 73%, while ‘other’ dropped from 33% to 11% (P = 0.036). Between eras, the CI of IM did not significantly increase in SCT (P = 0.27) or SOT (P = 0.30), and patterns of anatomic location (P = 0.122) and surgical debridement (P = 0.200) were similar. Significantly more patients received amphotericin‐echinocandin combination therapy in E2 (31% vs. 5%, P = 0.01); however, 90‐day survival did not improve (54% vs. 59%, P = 0.67). Since 2003, the rise of IM reflects increasing numbers at risk, not prior use of voriconazole. Frequent combination of anti‐fungal therapy has not improved survival.