Jeremy Lewin

Facilitating a culture of responsible and effective sharing of cancer genome data.

Rapid and affordable tumor molecular profiling has led to an explosion of clinical and genomic data poised to enhance the diagnosis, prognostication and treatment of cancer. A critical point has now been reached at which the analysis and storage of annotated clinical and genomic information in unconnected silos will stall the advancement of precision cancer care. Information systems must be harmonized to overcome the multiple technical and logistical barriers to data sharing. Against this backdrop, the Global Alliance for Genomic Health (GA4GH) was established in 2013 to create a common framework that enables responsible, voluntary and secure sharing of clinical and genomic data. This Perspective from the GA4GH Clinical Working Group Cancer Task Team highlights the data-aggregation challenges faced by the field, suggests potential collaborative solutions and describes how GA4GH can catalyze a harmonized data-sharing culture.

Development of Fibroblast Growth Factor Receptor Inhibitors: Kissing Frogs to Find a Prince?

Fibroblast growth factor (FGF) and fibroblast growth factor receptor (FGFR) are important regulators of a variety of biologic functions, including cellular proliferation, differentiation, migration, angiogenesis, wound healing, and survival. The FGF protein family comprises 18 ligands that signal through four transmembrane tyrosine kinase receptors (FGFR1-4). A fifth receptor, FGFR5 (or FGFRL1), can bind FGF, but it lacks a tyrosine kinase domain. Dysregulation of FGFR has been implicated in the development of many neoplasms and can occur through a variety of mechanisms, including gain-of-function mutations with constitutive kinase activation, chromosomal translocations with ligand-independent signaling, altered splicing, and gene amplification, which leads to receptor overexpression. Abnormalities in FGF and the FGFR pathway have been associated with progression of a wide spectrum of malignancies including myeloma, breast, endometrial, genitourinary, and gastric cancers. For example, amplification of the 8p12 loci which codes FGFR1 is detected in approximately 10% of breast cancers, FGFR2-activating mutations and amplifications are seen in 12% of endometrial cancers, and FGFR3 mutations can be seen in approximately 12% of bladder cancers. In addition, FGFs have been implicated in tumor angiogenesis and may mediate drug resistance to both conventional chemotherapy and anti–vascular endothelial growth factor (VEGF) therapy. Initial efforts in targeting FGFRs with small-molecule tyrosine kinase inhibitors (TKIs) have been tempered by challenges in the drug development process, which illustrates the complexities of developing drugs that target uncommon genomic alterations in tumors, as well as poor tolerability mainly related to nonspecificity and off-target effects. Multiple pharmaceutical companies are at different stages of pursuing FGFR blockade, mostly using small-molecule TKIs, but other approaches using monoclonal anti-FGFR antibodies and FGF trapping molecules are also being investigated. Early trials involved nonselective multitargeted TKIs that exhibit only modest bioactivity against FGFR and have wide-spectrum off-target inhibition against other tyrosine kinases, including VEGF receptors (VEGFRs). For example, dovitinib (TKI258) showed activity against FGFR1-3, VEGFR1-3, PDGFR-B, FLT-3, KIT, RET, and CSF1R, and lucitanib (E3810) is a potent inhibitor of VEGFR1-3, FGFR1-2, and CSF1R. Other nonselective FGFR inhibitors have been investigated (eg, nintedanib [BIBF1120], ponatinib [AP24534], brivanib [BMS-582664], lenvatinib [E7080], ENMD-2076, and orantinib [TSU-68]) and although they have some bioactivity against FGFR, their toxicity profiles (eg, hypertension and proteinuria) have been largely related to VEGFR inhibition. More recently, selective potent FGFR TKIs (eg, JNJ42756493, BGJ398, AZD4547, LY287445, and TAS120) are being investigated with high in vitro kinase activity and specificity against FGFR1, FGFR2, and FGFR3 (enzymatic concentration that causes 50% inhibition [IC50], 10 nmol/L) in the hope of having a more tolerable safety profile by reducing off-target effects. Of interest, selective FGFR inhibitors cause blockade of FGF23 release from the bone, acting as an on-target effect in normal tissues. The resultant modulations of serum phosphate, FGF23, and vitamin D could potentially be used as biomarkers of effective FGFR inhibition. The multitude of companies investigating FGFR inhibition need to be put in context with the difficulties of identifying patients with FGFR aberrations. In an early trial of dovitinib, molecular screening failure rates were high, and of the 243 patients with breast cancer who had their tumor samples analyzed for FGFR1 aberrations, only 25 were eligible for the FGFR1 amplified cohort. Similarly, in a study by Helsten et al that used Clinical Laboratory Improvement Amendments–approved next generation sequencing, only 343 of 4,853 patients who underwent molecular screening carried FGFR aberrations, primarily amplifications and activating missense mutations. Given limited resources and patients, the early drug development pathway for uncommon molecular aberrations should be scrutinized to determine whether multiple competing first-generation compounds should even be developed or whether investigation of this pathway should move forward only when more selective and potent lead candidates with minimal offtarget effects are identified. Having multiple first-generation compounds that ultimately undergo attrition is an inefficient process for advancing targeted therapeutics. It is in this context that Tabernero et al reported the results of a phase I trial of the selective FGFR inhibitor JNJ-42756493 in patients with advanced solid tumors. JNJ-42756493 is a potent, oral, panFGFR inhibitor with IC50 values in the low nanomolar ranges against FGFR1-4 that exhibits minimal activity against other kinases, including VEGFR. Although the initial recommended phase II dose (RP2D) of single-agent JNJ-42756493 was 9 mg per day, the authors declared 10 mg administered on an intermittent schedule (7 days on/7 days off) as the ultimate RP2D on the basis of its more favorable toxicity profile and the attainment of biologic relevance via pharmacokinetic simulation of preclinical efficacy data. Although blood-based pharmacodynamic assessment (eg, hyperphosphatemia) supported their RP2D, these are not validated biomarkers of target inhibition in the tumor or JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 33 NUMBER 30 OCTOBER 2

The positive effect of a dedicated adolescent and young adult fertility program on the rates of documentation of therapy-associated infertility risk and fertility preservation options

PurposeMinimal data exist regarding documentation of therapy-associated infertility risk (IR) and fertility preservation (FP) options during the initial oncology consultation prior to systemic therapy. This study investigated factors affecting IR/FP documentation and assessed the effect of implementation of an Adolescent and Young Adult (AYA) program on documentation rates.MethodsA retrospective review of charts of patients receiving gonadotoxic therapy was undertaken for documentation of IR/FP pre- and post-implementation of an AYA program. Change in documentation rates was assessed using univariate and multiple logistic regression.ResultsA total of 173 charts were reviewed. On univariate analysis, IR/FP documentation was less likely if patients had metastatic disease (Pxa0<xa00.01, Pxa0<xa00.01), by tumor type (Pxa0<xa00.01, Pxa0<xa00.01), received less intensive chemotherapy (Pxa0=xa00.03, Pxa0=xa00.06), were older (Pxa0=xa00.14, Pxa0<xa00.01), had more children (Pxa0<xa00.01, Pxa0<xa00.01), or lacked AYA program involvement (Pxa0<xa00.01, Pxa0<xa00.01). FP discussion was more common in males (Pxa0=xa00.02). On multivariable analysis, more children (Pxa0=xa00.01, Pxa0=xa00.03), older age (Pxa0<xa00.01, Pxa0<xa00.01), tumor type (Pxa0<xa00.01, Pxa0=xa00.01), stage (Pxa0=xa00.02, NS), relationship (Pxa0=xa00.03, NS), and lack of AYA involvement (Pxa0<xa00.01, Pxa0<xa00.01) were associated with lower rates of IR/FP documentation. Following AYA program implementation, IR/FP rates increased from 56% (CI 46–65%) to 85% (CI 74–92%, Pxa0<xa00.01) and 54% (CI 45–64%) to 86% (CI 75–93%, Pxa0<xa00.01), respectively. The effect of AYA program implementation on IR/FP documentation was most noticeable in leukemia, lymphoma, and breast groups (Pxa0<xa00.01).ConclusionsImplementing an AYA consultation service at an adult cancer institution had a positive effect on the rates of IR/FP documentation. Specific programming can improve service delivery to AYA cancer patients, and fertility counseling should be integrated for patients undergoing gonadotoxic therapy.

Identifying actionable variants using next generation sequencing in patients with a historical diagnosis of undifferentiated pleomorphic sarcoma

There are limited data regarding the molecular characterization of undifferentiated pleomorphic sarcomas (UPS; formerly malignant fibrous histiocytoma). This study aimed to investigate the utility of next generation sequencing (NGS) in UPS to identify subsets of patients who harbour actionable mutations. Patients diagnosed with UPS underwent pathological re‐evaluation by a pathologist specializing in sarcoma. Tumor DNA was isolated from archived fresh frozen tissue samples and genotyped using NGS with the Illumina MiSeq TruSeq Amplicon Cancer Panel (48 genes, 212 amplicons). In total, 95 patients initially classified with UPS were identified. Following pathology re‐review the histological subtypes were reclassified to include: Myxofibrosarcoma (MFS, N = 44); UPS(N = 18); and Others (N = 27; including undifferentiated spindle cell sarcoma (N = 15) and dedifferentiated liposarcoma (N = 6)). Seven cases were excluded from further analysis for other reasons. Baseline demographics of the finalized cohort (N = 88) showed a median age of 66 years (32–95), primarily with stage I–III disease (92%) and high‐grade (86%) lesions. Somatic mutations were identified in 31 cases (35%)(Total mutationsu2009=u200936: solitary mutation(n = 27); two mutations( =n = 3); three mutations(n = 1)). The most commonly identified mutations were in TP53 (nu2009=u200924), ATM (n = 3) and PIK3CA (nu2009=u20092). Three of 43 patients with MFS and one of 18 patients with UPS had clinically relevant mutations, mainly related to biomarkers of prediction of response; however few had targetable driver mutations. Somatic mutation status did not influence disease free or overall survival. Based on the small number of clinically relevant mutations, these data do not support the routine use of targeted NGS panels outside of research protocols in UPS.

Towards a global cancer knowledge network: dissecting the current international cancer genomic sequencing landscape

Background While next generation sequencing has enhanced our understanding of the biological basis of malignancy, current knowledge on global practices for sequencing cancer samples is limited. To address this deficiency, we developed a survey to provide a snapshot of current sequencing activities globally, identify barriers to data sharing and use this information to develop sustainable solutions for the cancer research community. Methods A multi-item survey was conducted assessing demographics, clinical data collection, genomic platforms, privacy/ethics concerns, funding sources and data sharing barriers for sequencing initiatives globally. Additionally, respondents were asked as to provide the primary intent of their initiative (clinical diagnostic, research or combination). Results Of 107 initiatives invited to participate, 59 responded (response rateu2009=u200955%). Whole exome sequencing (Pu2009=u20090.03) and whole genome sequencing (Pu2009=u20090.01) were utilized less frequently in clinical diagnostic than in research initiatives. Procedures to identify cancer-specific variants were heterogeneous, with bioinformatics pipelines employing different mutation calling/variant annotation algorithms. Measurement of treatment efficacy varied amongst initiatives, with time on treatment (57%) and RECIST (53%) being the most common; however, other parameters were also employed. Whilst 72% of initiatives indicated data sharing, its scope varied, with a number of restrictions in place (e.g. transfer of raw data). The largest perceived barriers to data harmonization were the lack of financial support (Pu2009<u20090.01) and bioinformatics concerns (e.g. lack of interoperability) (Pu2009=u20090.02). Capturing clinical data was more likely to be perceived as a barrier to data sharing by larger initiatives than by smaller initiatives (Pu2009=u20090.01). Conclusions These results identify the main barriers, as perceived by the cancer sequencing community, to effective sharing of cancer genomic and clinical data. They highlight the need for greater harmonization of technical, ethical and data capture processes in cancer sample sequencing worldwide, in order to support effective and responsible data sharing for the benefit of patients.BackgroundnWhile next generation sequencing has enhanced our understanding of the biological basis of malignancy, current knowledge on global practices for sequencing cancer samples is limited. To address this deficiency, we developed a survey to provide a snapshot of current sequencing activities globally, identify barriers to data sharing and use this information to develop sustainable solutions for the cancer research community.nnnMethodsnA multi-item survey was conducted assessing demographics, clinical data collection, genomic platforms, privacy/ethics concerns, funding sources and data sharing barriers for sequencing initiatives globally. Additionally, respondents were asked as to provide the primary intent of their initiative (Clinical Diagnostic, Research or Combination).nnnResultsnOf 107 initiatives invited to participate, 59 responded (response rate=55%). Whole Exome Sequencing ( p =0.03) and Whole Genome Sequencing ( pu2009=u2009 0.01), were utilized less frequently in Clinical Diagnostic than in Research initiatives. Procedures to identify cancer-specific variants were heterogeneous, with bioinformatics pipelines employing different mutation calling/variant annotation algorithms. Measurement of treatment efficacy varied amongst initiatives, with time on treatment (57%) and RECIST (53%) being the most common however; other parameters were also employed. Whilst 72% of initiatives indicated data sharing, its scope varied, with a number of restrictions in place (e.g. transfer of raw data). The largest perceived barriers to data harmonisation were the lack of financial support ( pu2009<u2009 0.01) and bioinformatics concerns (e.g. lack of interoperability)( pu2009=u2009 0.02). Capturing clinical data was more likely to be perceived as a barrier to data sharing by larger initiatives than by smaller initiatives ( pu2009=u2009 0.01).nnnConclusionsnThese results identify the main barriers, as perceived by the cancer sequencing community, to effective sharing of cancer genomic and clinical data. They highlight the need for greater harmonisation of technical, ethical and data capture processes in cancer sample sequencing worldwide, in order to support effective and responsible data sharing for the benefit of patients.

Shrinking Hepatic Hemangiomas in a Patient Treated for Metastatic Germ Cell Tumor

Metastatic germ cell tumors are treated with combination cisplatin chemotherapy, and decisions regarding the optimal number of cycles to deliver are determined by pathologic findings, site of origin, volume of disease, and degree of marker elevation using the International Germ Cell Cancer Collaborative Group Prognostic Risk Classification (IGCCCG). Patients with nonseminoma GCT with liver metastasis are considered poor risk according to the IGCCCG classification and treated with 4 cycles of platinumbased combination chemotherapy In the present case report, we describe a patient with a good-risk primary retroperitoneal GCT and liver lesions consistent with hemangiomas. Our patient received 3 cycles of bleomycin, etoposide, and cisplatin and had a chemoresponsive hepatic hemangioma that mimicked a liver metastasis. The primary tumor and liver lesions both shrank with systemic therapy, potentially from the effect of bleomycin. Chemoresponsive hepatic hemangiomas can mimic the response of liver metastases to therapy, making the ability to distinguish between the 2 difficult.

Second-line Chemotherapy in Older Patients With Metastatic Urothelial Carcinoma: Pooled Analysis of 10 Second-line Studies

Background Older patients with metastatic urothelial carcinoma (UC) are under‐represented in clinical trials, and data regarding outcomes for second‐line therapy is limited. Materials and Methods Individual data for patients with metastatic UC, aged ≥ 70 years, were pooled from 10 second‐line studies. The influence of potential prognostic factors on overall survival (OS) was assessed via univariate and multivariate Cox regression analysis. Results In total, 102 patients were included; the median age was 74.0 years (range, 70‐88 years). Second‐line chemotherapy was single‐agent in 42 (41%) patients and combination regimens in 60 (59%) patients. Median progression‐free and OS were 4.3 and 9.7 months, respectively. In multivariate analysis, age > 75 years, Eastern Cooperative Oncology Group performance status ≥ 1, serum hemoglobin < 10 g/dL, and non‐lymph node only metastasis predicted inferior OS. Median OS for patients with 0, 1, 2, and ≥ 3 adverse factors was unreached, 15.5, 9.8, and 4.8 months, respectively (P < .001). There was no difference in OS between patients treated with single‐agent or combination chemotherapy. Combination regimens were associated with higher occurrences of any ≥ grade 2 toxicity (80% vs. 38%; P < .001), ≥ grade 2 hematologic (78% vs. 12%; P < .001), and ≥ grade 2 gastrointestinal toxicity (36% vs. 7%; P < .001). Conclusion In this pooled analysis of older patients with metastatic UC, combination chemotherapy for second‐line treatment was associated with greater toxicity without improvement in OS. Eastern Cooperative Oncology Group performance status ≥1, serum hemoglobin < 10 g/dL, and age > 75 years predicted worse survival, whereas isolated lymph node metastasis predicted a favorable outcome. Micro‐Abstract We pooled individual data of older patients (≥ 70 years) from 10 studies that assessed second‐line chemotherapy for metastatic urothelial carcinoma. A prognostic model that separates the patients into 4 groups with significant differences in survival outcomes was developed. Furthermore, second‐line combination regimens, compared with single‐agent chemotherapy, were associated with increased toxicity without improved survival.

A New Model to Predict Benign Histology in Residual Retroperitoneal Masses After Chemotherapy in Nonseminoma

BACKGROUNDnPostchemotherapy retroperitoneal lymph node dissection (pcRPLND) is indicated in testicular cancer patients with normalised or plateaued serum tumour markers and residual retroperitoneal lesions >1cm. Challenges remain in predicting postchemotherapy residual mass (pcRM) histology, which may lead to unnecessary surgery.nnnOBJECTIVEnTo develop an accurate model to predict pcRM histology in patients with nonseminomatous germ cell tumours (NSGCTs).nnnDESIGN, SETTING, AND PARTICIPANTSnA retrospective review of 335 patients undergoing pcRPLND for metastatic NSGCTs to develop a model to predict benign histology in retroperitoneal pcRM. Our model was compared with others and externally validated.nnnINTERVENTIONnChemotherapy and pcRPLND.nnnOUTCOME MEASUREMENTS AND STATISTICAL ANALYSISnMultivariable logistic regression to evaluate the presence of benign histology, and fractional polynomials to allow for a nonlinear association between continuous variables and the outcome. The final Princess Margaret model (PMM) was selected based on the number of variables used, reliability, and discriminative capacity to predict benign pcRM.nnnRESULTS AND LIMITATIONSnPMM included the presence of teratoma in the orchiectomy, prechemotherapy α-fetoprotein, prechemotherapy mass size, and change in mass size during chemotherapy. Model specificity was 99.3%. Compared with Vergouwe et als model, PMM had significantly better accuracy (C statistic 0.843 vs 0.783). PMM appropriately identified a larger number of patients for whom pcRPLND can safely be avoided (13.9% vs 0%). Validated in external cohorts, the model retained high discrimination (C statistic 0.88 and 0.80). Larger and prospective studies are needed to further validate this model.nnnCONCLUSIONSnOur clinical model, externally validated, showed improved discriminative ability in predicting pcRM histology when compared with other models. The higher accuracy and reduced number of variables make this a novel and appealing model to use for patient counselling and treatment strategies.nnnPATIENT SUMMARYnPrincess Margaret model accurately predicted postchemotherapy benign histology. These results might have clinical impact by avoiding unnecessary retroperitoneal lymph node dissection and consequently changing the paradigm of advanced testicular cancer treatment.

Response to Immune Checkpoint Inhibition in Two Patients with Alveolar Soft Part Sarcoma

Two patients with ASPS responded to immune checkpoint inhibition. Genomic analysis of a larger group of patients demonstrated molecular mismatch repair deficiency signatures in 71% of patients. Immune checkpoint blockade may be a useful therapy for ASPS. Alveolar soft-part sarcoma (ASPS) is a morphologically distinctive mesenchymal tumor characterized by a canonical ASPL–TFE3 fusion product. In the metastatic setting, standard cytotoxic chemotherapies are typically ineffective. Studies have suggested modest clinical response to multitargeted receptor tyrosine kinase inhibitors. Here, we report sustained partial responses in two patients with immune checkpoint inhibition treated with either durvalumab (anti–PD-L1) alone or in combination with tremelimumab (anti–CTLA-4), which appeared unrelated to tumor immune infiltrates or mutational burden. Genomic analysis of these patients, and other cases of ASPS, demonstrated molecular mismatch-repair deficiency signatures. These findings suggest that immune checkpoint blockade may be a useful therapeutic strategy for ASPS. Cancer Immunol Res; 6(9); 1001–7. ©2018 AACR.

Tumor necrosis and clinical outcomes following neoadjuvant therapy in soft tissue sarcoma: A systematic review and meta-analysis

BACKGROUNDnThe prognostic role of tumor necrosis following neoadjuvant therapy is established in bone sarcomas but remains unclear in soft tissue sarcomas (STS).nnnMETHODSnWe searched MEDLINE, MEDLINE in progress, EMBASE and Cochrane to identify studies that investigated neoadjuvant therapy in STS. Studies were required to report survival data based on extent of necrosis, or provided individual patient data allowing estimation thereof. Hazard ratios (HR) for relapse-free (RFS) and overall survival (OS) and odds ratios (OR) for recurrence at 3u202fyears and for death at 5u202fyears were pooled in a random effect meta-analysis. Associations between patient characteristics and attainment of ≥90% necrosis were explored.nnnRESULTSn21 studies comprising 1663 patients were included. Extremity tumors were most common (nu202f=u202f1554; 93%). Induction regimens included chemotherapy with radiotherapy (nu202f=u202f924; 56%), chemotherapy alone (nu202f=u202f412; 25%), radiotherapy alone (nu202f=u202f78; 5%), isolated limb perfusion (ILP) (nu202f=u202f231; 14%), and targeted therapy/radiotherapy (nu202f=u202f18; 1%). Patients with <90% necrosis had higher hazard of recurrence (hazard ratio [HR] 1.47; 95% CI: 1.06-2.04; pu202f=u202f0.02) and death (HR 1.86; 95% CI: 1.41-2.46; pu202f<u202f0.001). Risk of recurrence at 3u202fyears (ORu202f=u202f3.35; 95% CI: 2.27-4.92; pu202f<u202f0.001) and of death at 5u202fyears (OR 2.60; 95% CI: 1.59-4.26; pu202f<u202f0.001) were similarly increased. Compared to other modalities, ILP was associated with higher odds of achieving ≥90% necrosis (OR 12.1; 95% CI: 3.69-39.88; pu202f<u202f0.001).nnnCONCLUSIONnTumour necrosis <90% following neoadjuvant therapy is associated with increased recurrence risk and inferior OS in patients with STS.