Multi-omic machine learning predictor of breast cancer therapy response.

Breast cancers are complex ecosystems of malignant cells and the tumour microenvironment1. The composition of these tumour ecosystems and interactions within them contribute to responses to cytotoxic therapy2. Efforts to build response predictors have not incorporated this knowledge. We collected clinical, digital pathology, genomic and transcriptomic profiles of pre-treatment biopsies of breast tumours from 168 patients treated with chemotherapy with or without HER2 (encoded by ERBB2)-targeted therapy before surgery. Pathology end points (complete response or residual disease) at surgery3 were then correlated with multi-omic features in these diagnostic biopsies. Here we show that response to treatment is modulated by the pre-treated tumour ecosystem, and its multi-omics landscape can be integrated in predictive models using machine learning. The degree of residual disease following therapy is monotonically associated with pre-therapy features, including tumour mutational and copy number landscapes, tumour proliferation, immune infiltration and T cell dysfunction and exclusion. Combining these features into a multi-omic machine learning model predicted a pathological complete response in an external validation cohort (75 patients) with an area under the curve of 0.87. In conclusion, response to therapy is determined by the baseline characteristics of the totality of the tumour ecosystem captured through data integration and machine learning. This approach could be used to develop predictors for other cancers.

Optimising the diagnostic accuracy of First post-contrAst SubtracTed breast MRI (FAST MRI) through interpretation-training: a multicentre e-learning study, mapping the learning curve of NHS Breast Screening Programme (NHSBSP) mammogram readers using an enriched dataset.

BACKGROUND: Abbreviated breast MRI (FAST MRI) is being introduced into clinical practice to screen women with mammographically dense breasts or with a personal history of breast cancer. This study aimed to optimise diagnostic accuracy through the adaptation of interpretation-training. METHODS: A FAST MRI interpretation-training programme (short presentations and guided hands-on workstation teaching) was adapted to provide additional training during the assessment task (interpretation of an enriched dataset of 125 FAST MRI scans) by giving readers feedback about the true outcome of each scan immediately after each scan was interpreted (formative assessment). Reader interaction with the FAST MRI scans used developed software (RiViewer) that recorded reader opinions and reading times for each scan. The training programme was additionally adapted for remote e-learning delivery. STUDY DESIGN: Prospective, blinded interpretation of an enriched dataset by multiple readers. RESULTS: 43 mammogram readers completed the training, 22 who interpreted breast MRI in their clinical role (Group 1) and 21 who did not (Group 2). Overall sensitivity was 83% (95%CI 81-84%; 1994/2408), specificity 94% (95%CI 93-94%; 7806/8338), readers' agreement with the true outcome kappa = 0.75 (95%CI 0.74-0.77) and diagnostic odds ratio = 70.67 (95%CI 61.59-81.09). Group 1 readers showed similar sensitivity (84%) to Group 2 (82% p = 0.14), but slightly higher specificity (94% v. 93%, p = 0.001). Concordance with the ground truth increased significantly with the number of FAST MRI scans read through the formative assessment task (p = 0.002) but by differing amounts depending on whether or not a reader had previously attended FAST MRI training (interaction p = 0.02). Concordance with the ground truth was significantly associated with reading batch size (p = 0.02), tending to worsen when more than 50 scans were read per batch. Group 1 took a median of 56 seconds (range 8-47,466) to interpret each FAST MRI scan compared with 78 (14-22,830, p < 0.0001) for Group 2. CONCLUSIONS: Provision of immediate feedback to mammogram readers during the assessment test set reading task increased specificity for FAST MRI interpretation and achieved high diagnostic accuracy. Optimal reading-batch size for FAST MRI was 50 reads per batch. Trial registration (25/09/2019): ISRCTN16624917.

Digital pathology for reporting histopathology samples, including cancer screening samples - definitive evidence from a multisite study.

AIMS: To conduct a definitive multicentre comparison of digital pathology (DP) with light microscopy (LM) for reporting histopathology slides including breast and bowel cancer screening samples. METHODS: A total of 2024 cases (608 breast, 607 GI, 609 skin, 200 renal) were studied, including 207 breast and 250 bowel cancer screening samples. Cases were examined by four pathologists (16 study pathologists across the four speciality groups), using both LM and DP, with the order randomly assigned and 6 weeks between viewings. Reports were compared for clinical management concordance (CMC), meaning identical diagnoses plus differences which do not affect patient management. Percentage CMCs were computed using logistic regression models with crossed random-effects terms for case and pathologist. The obtained percentage CMCs were referenced to 98.3% calculated from previous studies. RESULTS: For all cases LM versus DP comparisons showed the CMC rates were 99.95% [95% confidence interval (CI) = 99.90-99.97] and 98.96 (95% CI = 98.42-99.32) for cancer screening samples. In speciality groups CMC for LM versus DP showed: breast 99.40% (99.06-99.62) overall and 96.27% (94.63-97.43) for cancer screening samples; [gastrointestinal (GI) = 99.96% (99.89-99.99)] overall and 99.93% (99.68-99.98) for bowel cancer screening samples; skin 99.99% (99.92-100.0); renal 99.99% (99.57-100.0). Analysis of clinically significant differences revealed discrepancies in areas where interobserver variability is known to be high, in reads performed with both modalities and without apparent trends to either. CONCLUSIONS: Comparing LM and DP CMC, overall rates exceed the reference 98.3%, providing compelling evidence that pathologists provide equivalent results for both routine and cancer screening samples irrespective of the modality used.

Residual cancer burden after neoadjuvant chemotherapy and long-term survival outcomes in breast cancer: a multicentre pooled analysis of 5161 patients.

BACKGROUND: Previous studies have independently validated the prognostic relevance of residual cancer burden (RCB) after neoadjuvant chemotherapy. We used results from several independent cohorts in a pooled patient-level analysis to evaluate the relationship of RCB with long-term prognosis across different phenotypic subtypes of breast cancer, to assess generalisability in a broad range of practice settings. METHODS: In this pooled analysis, 12 institutes and trials in Europe and the USA were identified by personal communications with site investigators. We obtained participant-level RCB results, and data on clinical and pathological stage, tumour subtype and grade, and treatment and follow-up in November, 2019, from patients (aged ≥18 years) with primary stage I-III breast cancer treated with neoadjuvant chemotherapy followed by surgery. We assessed the association between the continuous RCB score and the primary study outcome, event-free survival, using mixed-effects Cox models with the incorporation of random RCB and cohort effects to account for between-study heterogeneity, and stratification to account for differences in baseline hazard across cancer subtypes defined by hormone receptor status and HER2 status. The association was further evaluated within each breast cancer subtype in multivariable analyses incorporating random RCB and cohort effects and adjustments for age and pretreatment clinical T category, nodal status, and tumour grade. Kaplan-Meier estimates of event-free survival at 3, 5, and 10 years were computed for each RCB class within each subtype. FINDINGS: We analysed participant-level data from 5161 patients treated with neoadjuvant chemotherapy between Sept 12, 1994, and Feb 11, 2019. Median age was 49 years (IQR 20-80). 1164 event-free survival events occurred during follow-up (median follow-up 56 months [IQR 0-186]). RCB score was prognostic within each breast cancer subtype, with higher RCB score significantly associated with worse event-free survival. The univariable hazard ratio (HR) associated with one unit increase in RCB ranged from 1·55 (95% CI 1·41-1·71) for hormone receptor-positive, HER2-negative patients to 2·16 (1·79-2·61) for the hormone receptor-negative, HER2-positive group (with or without HER2-targeted therapy; p<0·0001 for all subtypes). RCB score remained prognostic for event-free survival in multivariable models adjusted for age, grade, T category, and nodal status at baseline: the adjusted HR ranged from 1·52 (1·36-1·69) in the hormone receptor-positive, HER2-negative group to 2·09 (1·73-2·53) in the hormone receptor-negative, HER2-positive group (p<0·0001 for all subtypes). INTERPRETATION: RCB score and class were independently prognostic in all subtypes of breast cancer, and generalisable to multiple practice settings. Although variability in hormone receptor subtype definitions and treatment across patients are likely to affect prognostic performance, the association we observed between RCB and a patient's residual risk suggests that prospective evaluation of RCB could be considered to become part of standard pathology reporting after neoadjuvant therapy. FUNDING: National Cancer Institute at the US National Institutes of Health.

Evaluating the effectiveness of abbreviated breast MRI (abMRI) interpretation training for mammogram readers: a multi-centre study assessing diagnostic performance, using an enriched dataset.

BACKGROUND: Abbreviated breast MRI (abMRI) is being introduced in breast screening trials and clinical practice, particularly for women with dense breasts. Upscaling abMRI provision requires the workforce of mammogram readers to learn to effectively interpret abMRI. The purpose of this study was to examine the diagnostic accuracy of mammogram readers to interpret abMRI after a single day of standardised small-group training and to compare diagnostic performance of mammogram readers experienced in full-protocol breast MRI (fpMRI) interpretation (Group 1) with that of those without fpMRI interpretation experience (Group 2). METHODS: Mammogram readers were recruited from six NHS Breast Screening Programme sites. Small-group hands-on workstation training was provided, with subsequent prospective, independent, blinded interpretation of an enriched dataset with known outcome. A simplified form of abMRI (first post-contrast subtracted images (FAST MRI), displayed as maximum-intensity projection (MIP) and subtracted slice stack) was used. Per-breast and per-lesion diagnostic accuracy analysis was undertaken, with comparison across groups, and double-reading simulation of a consecutive screening subset. RESULTS: 37 readers (Group 1: 17, Group 2: 20) completed the reading task of 125 scans (250 breasts) (total = 9250 reads). Overall sensitivity was 86% (95% confidence interval (CI) 84-87%; 1776/2072) and specificity 86% (95%CI 85-86%; 6140/7178). Group 1 showed significantly higher sensitivity (843/952; 89%; 95%CI 86-91%) and higher specificity (2957/3298; 90%; 95%CI 89-91%) than Group 2 (sensitivity = 83%; 95%CI 81-85% (933/1120) p < 0.0001; specificity = 82%; 95%CI 81-83% (3183/3880) p < 0.0001). Inter-reader agreement was higher for Group 1 (kappa = 0.73; 95%CI 0.68-0.79) than for Group 2 (kappa = 0.51; 95%CI 0.45-0.56). Specificity improved for Group 2, from the first 55 cases (81%) to the remaining 70 (83%) (p = 0.02) but not for Group 1 (90-89% p = 0.44), whereas sensitivity remained consistent for both Group 1 (88-89%) and Group 2 (83-84%). CONCLUSIONS: Single-day abMRI interpretation training for mammogram readers achieved an overall diagnostic performance within benchmarks published for fpMRI but was insufficient for diagnostic accuracy of mammogram readers new to breast MRI to match that of experienced fpMRI readers. Novice MRI reader performance improved during the reading task, suggesting that additional training could further narrow this performance gap.

Diagnostic pathways in multiple myeloma and their relationship to end organ damage: an analysis from the Tackling Early Morbidity and Mortality in Myeloma (TEAMM) trial.

Multiple myeloma is associated with significant early morbidity and mortality, with considerable end organ damage often present at diagnosis. The Tackling EArly Morbidity and Mortality in Multiple Myeloma (TEAMM) trial was used to evaluate routes to diagnosis in patients with myeloma and the relationship between diagnostic pathways, time to diagnosis and disease severity. A total of 915 participants were included in the study. Fifty-one per cent were diagnosed by direct referral from primary care to haematology; 29% were diagnosed via acute services and 20% were referred via other secondary care specialties. Patients diagnosed via other secondary care specialties had a longer diagnostic interval (median 120 days vs. 59 days) without an increase in features of severe disease, suggesting they had a relatively indolent disease. Marked intrahospital delay suggests possible scope for improvement. A quarter of those diagnosed through acute services reported >30 days from initial hospital consultation to haematology assessment. Participants diagnosed through acute services had poorer performance status (P < 0·0001) and higher burden of end organ damage (P < 0·0001) with no difference in the overall length of diagnostic pathway compared to those diagnosed by direct referral (median 59 days). This suggests that advanced disease in patients presenting through acute services predominantly reflects disease aggression.

A multicentre prospective feasibility study of carbon dye tattooing of biopsied axillary node and surgical localisation in breast cancer patients.

BACKGROUND: The primary aim of this prospective, multicentre feasibility study was to determine whether the biopsied axillary node can be marked using black carbon dye and successfully identified at the time of surgery. METHODS: We included breast cancer patients undergoing needle biopsy of the axillary node. The biopsied node was tattooed at the time of needle biopsy (fine needle aspiration or core biopsy) or at a separate visit with black carbon dye (Spot® or Black Eye™). Participants underwent primary surgery or neoadjuvant chemotherapy (NACT) and axillary surgery (SNB or ALND) as per routine care. RESULTS: 110 patients were included. Median age of the women was 59 (range 31-88) years. 48 (44%) underwent SNB and 62 (56%) ALND. Median volume of dye injected was 2.0 ml (range 0.2-4.2). Tattooed node was identified intra-operatively in 90 (82%) patients. The identification rate was higher (76 of 88, 86%) in the primary surgery group compared with NACT (14 of 22, 64%) (p = 0.03). Of those undergoing NACT, the identification rate was better in the patients undergoing SNB (3 of 4, 75%) compared with ALND (11 of 18, 61%) (p > 0.99). The tattooed node was the sentinel node in 78% (28 of 36) patients in the primary surgery group and 100% (3 of 3) in the NACT group. There was no learning curve for surgeons or radiologists. The identification rate did not vary with timing between dye injection and surgery (p = 0.56), body mass index (p = 0.62) or volume of dye injected (p = 0.25). CONCLUSION: It is feasible to mark the axillary node with carbon dye and identify it intra-operatively. ClinicalTrials.gov: NCT03640819.

Radiotherapy plus cisplatin or cetuximab in low-risk human papillomavirus-positive oropharyngeal cancer (De-ESCALaTE HPV): an open-label randomised controlled phase 3 trial.

BACKGROUND: The incidence of human papillomavirus (HPV)-positive oropharyngeal cancer, a disease affecting younger patients, is rapidly increasing. Cetuximab, an epidermal growth factor receptor inhibitor, has been proposed for treatment de-escalation in this setting to reduce the toxicity of standard cisplatin treatment, but no randomised evidence exists for the efficacy of this strategy. METHODS: We did an open-label randomised controlled phase 3 trial at 32 head and neck treatment centres in Ireland, the Netherlands, and the UK, in patients aged 18 years or older with HPV-positive low-risk oropharyngeal cancer (non-smokers or lifetime smokers with a smoking history of <10 pack-years). Eligible patients were randomly assigned (1:1) to receive, in addition to radiotherapy (70 Gy in 35 fractions), either intravenous cisplatin (100 mg/m2 on days 1, 22, and 43 of radiotherapy) or intravenous cetuximab (400 mg/m2 loading dose followed by seven weekly infusions of 250 mg/m2). The primary outcome was overall severe (grade 3-5) toxicity events at 24 months from the end of treatment. The primary outcome was assessed by intention-to-treat and per-protocol analyses. This trial is registered with the ISRCTN registry, number ISRCTN33522080. FINDINGS: Between Nov 12, 2012, and Oct 1, 2016, 334 patients were recruited (166 in the cisplatin group and 168 in the cetuximab group). Overall (acute and late) severe (grade 3-5) toxicity did not differ significantly between treatment groups at 24 months (mean number of events per patient 4·8 [95% CI 4·2-5·4] with cisplatin vs 4·8 [4·2-5·4] with cetuximab; p=0·98). At 24 months, overall all-grade toxicity did not differ significantly either (mean number of events per patient 29·2 [95% CI 27·3-31·0] with cisplatin vs 30·1 [28·3-31·9] with cetuximab; p=0·49). However, there was a significant difference between cisplatin and cetuximab in 2-year overall survival (97·5% vs 89·4%, hazard ratio 5·0 [95% CI 1·7-14·7]; p=0·001) and 2-year recurrence (6·0% vs 16·1%, 3·4 [1·6-7·2]; p=0·0007). INTERPRETATION: Compared with the standard cisplatin regimen, cetuximab showed no benefit in terms of reduced toxicity, but instead showed significant detriment in terms of tumour control. Cisplatin and radiotherapy should be used as the standard of care for HPV-positive low-risk patients who are able to tolerate cisplatin. FUNDING: Cancer Research UK.

6 versus 12 months of adjuvant trastuzumab for HER2-positive early breast cancer (PERSEPHONE): 4-year disease-free survival results of a randomised phase 3 non-inferiority trial.

BACKGROUND: Adjuvant trastuzumab significantly improves outcomes for patients with HER2-positive early breast cancer. The standard treatment duration is 12 months but shorter treatment could provide similar efficacy while reducing toxicities and cost. We aimed to investigate whether 6-month adjuvant trastuzumab treatment is non-inferior to the standard 12-month treatment regarding disease-free survival. METHODS: This study is an open-label, randomised phase 3 non-inferiority trial. Patients were recruited from 152 centres in the UK. We randomly assigned patients with HER2-positive early breast cancer, aged 18 years or older, and with a clear indication for chemotherapy, by a computerised minimisation process (1:1), to receive either 6-month or 12-month trastuzumab delivered every 3 weeks intravenously (loading dose of 8 mg/kg followed by maintenance doses of 6 mg/kg) or subcutaneously (600 mg), given in combination with chemotherapy (concurrently or sequentially). The primary endpoint was disease-free survival, analysed by intention to treat, with a non-inferiority margin of 3% for 4-year disease-free survival. Safety was analysed in all patients who received trastuzumab. This trial is registered with EudraCT (number 2006-007018-39), ISRCTN (number 52968807), and ClinicalTrials.gov (number NCT00712140). FINDINGS: Between Oct 4, 2007, and July 31, 2015, 2045 patients were assigned to 12-month trastuzumab treatment and 2044 to 6-month treatment (one patient was excluded because they were double randomised). Median follow-up was 5·4 years (IQR 3·6-6·7) for both treatment groups, during which a disease-free survival event occurred in 265 (13%) of 2043 patients in the 6-month group and 247 (12%) of 2045 patients in the 12-month group. 4-year disease-free survival was 89·4% (95% CI 87·9-90·7) in the 6-month group and 89·8% (88·3-91·1) in the 12-month group (hazard ratio 1·07 [90% CI 0·93-1·24], non-inferiority p=0·011), showing non-inferiority of the 6-month treatment. 6-month trastuzumab treatment resulted in fewer patients reporting severe adverse events (373 [19%] of 1939 patients vs 459 [24%] of 1894 patients, p=0·0002) or stopping early because of cardiotoxicity (61 [3%] of 1939 patients vs 146 [8%] of 1894 patients, p<0·0001). INTERPRETATION: We have shown that 6-month trastuzumab treatment is non-inferior to 12-month treatment in patients with HER2-positive early breast cancer, with less cardiotoxicity and fewer severe adverse events. These results support consideration of reduced duration trastuzumab for women at similar risk of recurrence as to those included in the trial. FUNDING: UK National Institute for Health Research, Health Technology Assessment Programme.

Comparison of Whole-Body MRI, CT, and Bone Scintigraphy for Response Evaluation of Cancer Therapeutics in Metastatic Breast Cancer to Bone.

Background CT and bone scintigraphy have limitations in evaluating systemic anticancer therapy (SACT) response in bone metastases from metastatic breast cancer (MBC). Purpose To evaluate whether whole-body MRI enables identification of progressive disease (PD) earlier than CT and bone scintigraphy in bone-only MBC. Materials and Methods This prospective study evaluated participants with bone-only MBC between May 2016 and January 2019 (ClinicalTrials.gov identifier: NCT03266744). Participants were enrolled at initiation of first or subsequent SACT based on standard CT and bone scintigraphy imaging. Baseline whole-body MRI was performed within 2 weeks of entry; those with extraosseous disease were excluded. CT and whole-body MRI were performed every 12 weeks until definitive PD was evident with one or both modalities. In case of PD, bone scintigraphy was used to assess for bone disease progression. Radiologists independently interpreted images from CT, whole-body MRI, or bone scintigraphy and were blinded to results with the other modalities. Systematic differences in performance between modalities were analyzed by using the McNemar test. Results Forty-five participants (mean age, 60 years ± 13 [standard deviation]; all women) were evaluated. Median time on study was 36 weeks (range, 1-120 weeks). Two participants were excluded because of unequivocal evidence of liver metastases at baseline whole-body MRI, two participants were excluded because they had clinical progression before imaging showed PD, and one participant was lost to follow-up. Of the 33 participants with PD at imaging, 67% (22 participants) had PD evident at whole-body MRI only and 33% (11 participants) had PD at CT and whole-body MRI concurrently; none had PD at CT only (P < .001, McNemar test). There was only slight agreement between whole-body MRI and CT (Cohen κ, 0.15). PD at bone scintigraphy was reported in 50% of participants (13 of 26) with bone progression at CT and/or whole-body MRI (P < .001, McNemar test). Conclusion Whole-body MRI enabled identification of progressive disease before CT in most participants with bone-only metastatic breast cancer. Progressive disease at bone scintigraphy was evident in only half of participants with bone progression at whole-body MRI. © RSNA, 2020 Online supplemental material is available for this article.

Levofloxacin prophylaxis in patients with newly diagnosed myeloma (TEAMM): a multicentre, double-blind, placebo-controlled, randomised, phase 3 trial.

BACKGROUND: Myeloma causes profound immunodeficiency and recurrent, serious infections. Around 5500 new cases of myeloma are diagnosed per year in the UK, and a quarter of patients will have a serious infection within 3 months of diagnosis. We aimed to assess whether patients newly diagnosed with myeloma benefit from antibiotic prophylaxis to prevent infection, and to investigate the effect on antibiotic-resistant organism carriage and health care-associated infections in patients with newly diagnosed myeloma. METHODS: TEAMM was a prospective, multicentre, double-blind, placebo-controlled randomised trial in patients aged 21 years and older with newly diagnosed myeloma in 93 UK hospitals. All enrolled patients were within 14 days of starting active myeloma treatment. We randomly assigned patients (1:1) to levofloxacin or placebo with a computerised minimisation algorithm. Allocation was stratified by centre, estimated glomerular filtration rate, and intention to proceed to high-dose chemotherapy with autologous stem cell transplantation. All investigators, patients, laboratory, and trial co-ordination staff were masked to the treatment allocation. Patients were given 500 mg of levofloxacin (two 250 mg tablets), orally once daily for 12 weeks, or placebo tablets (two tablets, orally once daily for 12 weeks), with dose reduction according to estimated glomerular filtration rate every 4 weeks. Follow-up visits occurred every 4 weeks up to week 16, and at 1 year. The primary outcome was time to first febrile episode or death from all causes within the first 12 weeks of trial treatment. All randomised patients were included in an intention-to-treat analysis of the primary endpoint. This study is registered with the ISRCTN registry, number ISRCTN51731976, and the EU Clinical Trials Register, number 2011-000366-35. FINDINGS: Between Aug 15, 2012, and April 29, 2016, we enrolled and randomly assigned 977 patients to receive levofloxacin prophylaxis (489 patients) or placebo (488 patients). Median follow-up was 12 months (IQR 8-13). 95 (19%) first febrile episodes or deaths occurred in 489 patients in the levofloxacin group versus 134 (27%) in 488 patients in the placebo group (hazard ratio 0·66, 95% CI 0·51-0·86; p=0·0018. 597 serious adverse events were reported up to 16 weeks from the start of trial treatment (308 [52%] of which were in the levofloxacin group and 289 [48%] of which were in the placebo group). Serious adverse events were similar between the two groups except for five episodes (1%) of mostly reversible tendonitis in the levofloxacin group. INTERPRETATION: Addition of prophylactic levofloxacin to active myeloma treatment during the first 12 weeks of therapy significantly reduced febrile episodes and deaths compared with placebo without increasing health care-associated infections. These results suggest that prophylactic levofloxacin could be used for patients with newly diagnosed myeloma undergoing anti-myeloma therapy. FUNDING: UK National Institute for Health Research.

PET-CT Surveillance versus Neck Dissection in Advanced Head and Neck Cancer.

BACKGROUND: The role of image-guided surveillance as compared with planned neck dissection in the treatment of patients with squamous-cell carcinoma of the head and neck who have advanced nodal disease (stage N2 or N3) and who have received chemoradiotherapy for primary treatment is a matter of debate. METHODS: In this prospective, randomized, controlled trial, we assessed the noninferiority of positron-emission tomography-computed tomography (PET-CT)-guided surveillance (performed 12 weeks after the end of chemoradiotherapy, with neck dissection performed only if PET-CT showed an incomplete or equivocal response) to planned neck dissection in patients with stage N2 or N3 disease. The primary end point was overall survival. RESULTS: From 2007 through 2012, we recruited 564 patients (282 patients in the planned-surgery group and 282 patients in the surveillance group) from 37 centers in the United Kingdom. Among these patients, 17% had nodal stage N2a disease and 61% had stage N2b disease. A total of 84% of the patients had oropharyngeal cancer, and 75% had tumor specimens that stained positive for the p16 protein, an indicator that human papillomavirus had a role in the causation of the cancer. The median follow-up was 36 months. PET-CT-guided surveillance resulted in fewer neck dissections than did planned dissection surgery (54 vs. 221); rates of surgical complications were similar in the two groups (42% and 38%, respectively). The 2-year overall survival rate was 84.9% (95% confidence interval [CI], 80.7 to 89.1) in the surveillance group and 81.5% (95% CI, 76.9 to 86.3) in the planned-surgery group. The hazard ratio for death slightly favored PET-CT-guided surveillance and indicated noninferiority (upper boundary of the 95% CI for the hazard ratio, <1.50; P=0.004). There was no significant difference between the groups with respect to p16 expression. Quality of life was similar in the two groups. PET-CT-guided surveillance, as compared with neck dissection, resulted in savings of £1,492 (approximately $2,190 in U.S. dollars) per person over the duration of the trial. CONCLUSIONS: Survival was similar among patients who underwent PET-CT-guided surveillance and those who underwent planned neck dissection, but surveillance resulted in considerably fewer operations and it was more cost-effective. (Funded by the National Institute for Health Research Health Technology Assessment Programme and Cancer Research UK; PET-NECK Current Controlled Trials number, ISRCTN13735240.).

25-hydroxyvitamin D serum levels in patients with high risk resected melanoma treated in an adjuvant bevacizumab trial.

BACKGROUND: Studies evaluating a relationship of vitamin D in patients with primary melanoma have consistently identified an inverse correlation with Breslow thickness, but an inconsistent impact on survival. Vitamin D in later stages of melanoma has been less studied. METHODS: Vitamin D was measured in serum from 341 patients with resected stage IIB-IIIC melanoma recruited to the AVAST-M adjuvant melanoma randomised trial, collected prior to randomisation, then at 3 and 12 months. Vitamin D levels were compared with patient demographics, known melanoma prognostic factors, disease-free interval (DFI) and overall survival (OS). RESULTS: A total of 73% patients had stage III melanoma, 32% were enroled (and therefore tested) >1 year after primary melanoma diagnosis. Median pre-randomisation vitamin D level was 56.5 (range 12.6-189.0 nmol/L). Vitamin D levels did not significantly vary over 12 months (p = 0.24). Individual pre-randomisation vitamin D levels did not differ significantly for Breslow thickness, tumour ulceration, or disease stage. Neither did pre-randomisation vitamin D predict for DFI (HR = 0.98 per 10 nmol/L increase; 95% confidence interval (CI) 0.93-1.04, p = 0.59) or OS (HR = 0.96 per 10 nmol/L increase, 95% CI 0.90-1.03, p = 0.31). For stage II patients, DFI improved with higher pre-randomisation vitamin D levels for those on bevacizumab (HR = 0.74 per 10 nmol nmol/L increase; 95% CI 0.56-0.97), but not for the observation arm (HR = 1.07 per 10 nmol/L increase; 95% CI 0.85-1.34). CONCLUSIONS: In this stage II/III melanoma cohort, vitamin D did not correlate with known prognostic markers, nor predict for DFI or OS, but there was some evidence of benefit for patients with stage II disease treated with bevacizumab.

Double Reading in Breast Cancer Screening: Cohort Evaluation in the CO-OPS Trial.

Purpose To investigate the effect of double readings by a second radiologist on recall rates, cancer detection, and characteristics of cancers detected in the National Health Service Breast Screening Program in England. Materials and Methods In this retrospective analysis, 805 206 women were evaluated through screening and diagnostic test results by extracting 1 year of routine data from 33 English breast screening centers. Centers used double reading of digital mammograms, with arbitration if there were discrepant interpretations. Information on reader decisions, with results of follow-up tests, were used to explore the effect of the second reader. The statistical tests used were the test for equality of proportions, the χ2 test for independence, and the t test. Results The first reader recalled 4.76% of women (38 295 of 805 206 women; 95% confidence interval [CI]: 4.71%, 4.80%). Two readers recalled 6.19% of women in total (49 857 of 805 206 women; 95% CI: 6.14%, 6.24%), but arbitration of discordant readings reduced the recall rate to 4.08% (32 863 of 805 206 women; 95% CI: 4.04%, 4.12%; P < .001). A total of 7055 cancers were detected, of which 627 (8.89%; 95% CI: 8.22%, 9.55%; P < .001) were detected by the second reader only. These additional cancers were more likely to be ductal carcinoma in situ (30.5% [183 of 600] vs 22.0% [1344 of 6114]; P < .001), and additional invasive cancers were smaller (mean size, 14.2 vs 16.7 mm; P < .001), had fewer involved nodes, and were likely to be lower grade. Conclusion Double reading with arbitration reduces recall and increases cancer detection compared with single reading. Cancers detected only by the second reader were smaller, of lower grade, and had less nodal involvement. © RSNA, 2018.

Adjuvant trastuzumab duration trials in HER2 positive breast cancer - what results would be practice-changing? Persephone investigator questionnaire prior to primary endpoint results.

BACKGROUND: Twelve months treatment is the current standard of care for adjuvant trastuzumab in patients with HER2 positive early breast cancer however the optimal duration is not known. Persephone is a non-inferiority randomised controlled trial comparing 6- to 12-months of trastuzumab. In this trial there will be a trade-off between a possible small decrease in disease-free survival (DFS) with 6-months and reduced cardiotoxicity and cost. METHODS: A structured questionnaire asked clinicians who had recruited patients into the Persephone trial about their prior beliefs with regards to the clinical effectiveness of trastuzumab and cardiotoxicity profile, in the comparison of 6- and 12-month durations. RESULTS: Fifty-one clinicians from 40 of the 152 Persephone sites completed the questionnaire. 30/50 responders (60%) believed that 6-months trastuzumab would give the same 4-year DFS rate as 12-months trastuzumab, with 21/50 (42%) holding this belief across all breast cancer subsets. In addition, 46/49 responders (94%) reported expecting to change their clinical practice to 6-months, with their prior beliefs (most commonly 85% 4-year DFS rate with 6-months) being greater than their lowest acceptable rate (most commonly 83% 4-year DFS rate with 6-months). Low levels of cardiotoxicity were expected with both 6 and 12-months trastuzumab, with the majority expecting lower levels with 6-months. With increasing hypothesised differences of cardiotoxicity rates between the two durations, significantly lower levels of 4-year DFS with 6-months trastuzumab were deemed acceptable (p < 0.0001). CONCLUSION: Most responders believe that 6-months trastuzumab is adequate, both overall and within each subset of breast cancer, and plan to change their clinical practice if the Persephone results support their prior belief. An individual patient meta-analysis of the duration trials would give greater precision to estimates of the differences in efficacy and toxicity, and adequate statistical power to establish a 2% level of non-inferiority for 6-months adjuvant trastuzumab.

Addition of gemcitabine to paclitaxel, epirubicin, and cyclophosphamide adjuvant chemotherapy for women with early-stage breast cancer (tAnGo): final 10-year follow-up of an open-label, randomised, phase 3 trial.

BACKGROUND: The tAnGo trial was designed to investigate the potential role of gemcitabine when added to anthracycline and taxane-containing adjuvant chemotherapy for early breast cancer. When this study was developed, gemcitabine had shown significant activity in metastatic breast cancer, and there was evidence of a favourable interaction with paclitaxel. METHODS: tAnGo was an international, open-label, randomised, phase 3 superiority trial that enrolled women aged 18 years or older with newly diagnosed, early-stage breast cancer who had a definite indication for chemotherapy, any nodal status, any hormone receptor status, Eastern Cooperative Oncology Group performance status of 0-1, and adequate bone marrow, hepatic, and renal function. Women were recruited from 127 clinical centres and hospitals in the UK and Ireland, and randomly assigned (1:1) to one of two treatment regimens: epirubicin, cyclophosphamide, and paclitaxel (four cycles of 90 mg/m2 intravenously administered epirubicin and 600 mg/m2 intravenously administered cyclophosphamide on day 1 every 3 weeks, followed by four cycles of 175 mg/m2 paclitaxel as a 3 h infusion on day 1 every 3 weeks) or epirubicin, cyclophosphamide, and paclitaxel plus gemcitabine (the same chemotherapy regimen as the other group, with the addition of 1250 mg/m2 gemcitabine to the paclitaxel cycles, administered intravenously as a 0·5 h infusion on days 1 and 8 every 3 weeks). Patients were randomly assigned by a central computerised deterministic minimisation procedure, with stratification by country, age, radiotherapy intent, nodal status, and oestrogen receptor and HER-2 status. The primary endpoint was disease-free survival and the trial aimed to detect 5% differences in 5-year disease-free survival between the treatment groups. Recruitment completed in 2004 and this is the final, intention-to-treat analysis. This trial is registered with EudraCT (2004-002927-41), ISRCTN (51146252), and ClinicalTrials.gov (NCT00039546). FINDINGS: Between Aug 22, 2001, and Nov 26, 2004, 3152 patients were enrolled and randomly assigned to epirubicin, cyclophosphamide, paclitaxel, and gemcitabine (gemcitabine group; n=1576) or to epirubicin, cyclophosphamide, and paclitaxel (control group; n=1576). 11 patients (six in the gemcitabine group and five in the control group) were ineligible because of pre-existing metastases and were therefore excluded from the analysis. At this protocol-specified final analysis (median follow-up 10 years [IQR 10-10]), 1087 disease-free survival events and 914 deaths had occurred. Disease-free survival did not differ significantly between the treatment groups at 10 years (65% [63-68] in the gemcitabine group vs 65% [62-67] in the control group), and median disease-free survival was not reached (adjusted hazard ratio 0·97 [95% CI 0·86-1·10], p=0·64). Toxicity, dose intensity, and a detailed safety substudy showed both regimens to be safe, deliverable, and tolerable. Grade 3 and 4 toxicities were reported at expected levels in both groups. The most common were neutropenia (527 [34%] of 1565 patients in the gemcitabine group vs 412 [26%] of 1567 in the control group), myalgia and arthralgia (207 [13%] vs 186 [12%]), fatigue (207 [13%] vs 152 [10%]), infection (202 [13%] vs 141 [9%]), vomiting (143 [9%] vs 108 [7%]), and nausea (132 [8%] vs 102 [7%]). INTERPRETATION: The addition of gemcitabine to anthracycline and taxane-based adjuvant chemotherapy at this dose and schedule confers no therapeutic advantage in terms of disease-free survival in early breast cancer, although it can cause increased toxicity. Therefore, gemcitabine has not been added to standard adjuvant chemotherapy in breast cancer for any subgroup. FUNDING: Cancer Research UK core funding for Clinical Trials Unit at the University of Birmingham, Eli Lilly, Bristol-Myers Squibb, and Pfizer.

Central pathology review with two-stage quality assurance for pathological response after neoadjuvant chemotherapy in the ARTemis Trial.

The ARTemis Trial tested standard neoadjuvant chemotherapy±bevacizumab in the treatment of HER2-negative early breast cancer. We compare data from central pathology review with report review and also the reporting behavior of the two central pathologists. Eight hundred women with HER2-negative early invasive breast cancer were recruited. Response to chemotherapy was assessed from local pathology reports for pathological complete response in breast and axillary lymph nodes. Sections from the original core biopsy and surgical excision were centrally reviewed by one of two trial pathologists blinded to the local pathology reports. Pathologists recorded response to chemotherapy descriptively and also calculated residual cancer burden. 10% of cases were double-reported to compare the central pathologists' reporting behavior. Full sample retrieval was obtained for 681 of the 781 patients (87%) who underwent surgery within the trial and were evaluable for pathological complete response. Four hundred and eighty-three (71%) were assessed by JSJT, and 198 (29%) were assessed by EP. Residual cancer burden calculations were possible in 587/681 (86%) of the centrally reviewed patients, as 94/681 (14%) had positive sentinel nodes removed before neoadjuvant chemotherapy invalidating residual cancer burden scoring. Good concordance was found between the two pathologists for residual cancer burden classes within the 65-patient quality assurance exercise (kappa 0.63 (95% CI: 0.57-0.69)). Similar results were obtained for the between-treatment arm comparison both from the report review and the central pathology review. For pathological complete response, report review was as good as central pathology review but for minimal residual disease, report review overestimated the extent of residual disease. In the ARTemis Trial central pathology review added little in the determination of pathological complete response but had a role in evaluating low levels of residual disease. Calculation of residual cancer burden was a simple and reproducible method of quantifying response to neoadjuvant chemotherapy as demonstrated by performance comparison of the two pathologists.

Value of Information Analysis of Multiparameter Tests for Chemotherapy in Early Breast Cancer: The OPTIMA Prelim Trial.

BACKGROUND: Precision medicine is heralded as offering more effective treatments to smaller targeted patient populations. In breast cancer, adjuvant chemotherapy is standard for patients considered as high-risk after surgery. Molecular tests may identify patients who can safely avoid chemotherapy. OBJECTIVES: To use economic analysis before a large-scale clinical trial of molecular testing to confirm the value of the trial and help prioritize between candidate tests as randomized comparators. METHODS: Women with surgically treated breast cancer (estrogen receptor-positive and lymph node-positive or tumor size ≥30 mm) were randomized to standard care (chemotherapy for all) or test-directed care using Oncotype DX™. Additional testing was undertaken using alternative tests: MammaPrintTM, PAM-50 (ProsignaTM), MammaTyperTM, IHC4, and IHC4-AQUA™ (NexCourse Breast™). A probabilistic decision model assessed the cost-effectiveness of all tests from a UK perspective. Value of information analysis determined the most efficient publicly funded ongoing trial design in the United Kingdom. RESULTS: There was an 86% probability of molecular testing being cost-effective, with most tests producing cost savings (range -£1892 to £195) and quality-adjusted life-year gains (range 0.17-0.20). There were only small differences in costs and quality-adjusted life-years between tests. Uncertainty was driven by long-term outcomes. Value of information demonstrated value of further research into all tests, with Prosigna currently being the highest priority for further research. CONCLUSIONS: Molecular tests are likely to be cost-effective, but an optimal test is yet to be identified. Health economics modeling to inform the design of a randomized controlled trial looking at diagnostic technology has been demonstrated to be feasible as a method for improving research efficiency.

A literature review of applied adaptive design methodology within the field of oncology in randomised controlled trials and a proposed extension to the CONSORT guidelines.

BACKGROUND: The application of adaptive design methodology within a clinical trial setting is becoming increasingly popular. However the application of these methods within trials is not being reported as adaptive designs hence making it more difficult to capture the emerging use of these designs. Within this review, we aim to understand how adaptive design methodology is being reported, whether these methods are explicitly stated as an 'adaptive design' or if it has to be inferred and to identify whether these methods are applied prospectively or concurrently. METHODS: Three databases; Embase, Ovid and PubMed were chosen to conduct the literature search. The inclusion criteria for the review were phase II, phase III and phase II/III randomised controlled trials within the field of Oncology that published trial results in 2015. A variety of search terms related to adaptive designs were used. RESULTS: A total of 734 results were identified, after screening 54 were eligible. Adaptive designs were more commonly applied in phase III confirmatory trials. The majority of the papers performed an interim analysis, which included some sort of stopping criteria. Additionally only two papers explicitly stated the term 'adaptive design' and therefore for most of the papers, it had to be inferred that adaptive methods was applied. Sixty-five applications of adaptive design methods were applied, from which the most common method was an adaptation using group sequential methods. CONCLUSIONS: This review indicated that the reporting of adaptive design methodology within clinical trials needs improving. The proposed extension to the current CONSORT 2010 guidelines could help capture adaptive design methods. Furthermore provide an essential aid to those involved with clinical trials.

Computational pathology of pre-treatment biopsies identifies lymphocyte density as a predictor of response to neoadjuvant chemotherapy in breast cancer.

BACKGROUND: There is a need to improve prediction of response to chemotherapy in breast cancer in order to improve clinical management and this may be achieved by harnessing computational metrics of tissue pathology. We investigated the association between quantitative image metrics derived from computational analysis of digital pathology slides and response to chemotherapy in women with breast cancer who received neoadjuvant chemotherapy. METHODS: We digitised tissue sections of both diagnostic and surgical samples of breast tumours from 768 patients enrolled in the Neo-tAnGo randomized controlled trial. We subjected digital images to systematic analysis optimised for detection of single cells. Machine-learning methods were used to classify cells as cancer, stromal or lymphocyte and we computed estimates of absolute numbers, relative fractions and cell densities using these data. Pathological complete response (pCR), a histological indicator of chemotherapy response, was the primary endpoint. Fifteen image metrics were tested for their association with pCR using univariate and multivariate logistic regression. RESULTS: Median lymphocyte density proved most strongly associated with pCR on univariate analysis (OR 4.46, 95 % CI 2.34-8.50, p < 0.0001; observations = 614) and on multivariate analysis (OR 2.42, 95 % CI 1.08-5.40, p = 0.03; observations = 406) after adjustment for clinical factors. Further exploratory analyses revealed that in approximately one quarter of cases there was an increase in lymphocyte density in the tumour removed at surgery compared to diagnostic biopsies. A reduction in lymphocyte density at surgery was strongly associated with pCR (OR 0.28, 95 % CI 0.17-0.47, p < 0.0001; observations = 553). CONCLUSIONS: A data-driven analysis of computational pathology reveals lymphocyte density as an independent predictor of pCR. Paradoxically an increase in lymphocyte density, following exposure to chemotherapy, is associated with a lack of pCR. Computational pathology can provide objective, quantitative and reproducible tissue metrics and represents a viable means of outcome prediction in breast cancer. TRIAL REGISTRATION: ClinicalTrials.gov NCT00070278 ; 03/10/2003.