Objectives
A convolutional neural network (CNN) was adapted to automatically detect early-stage nasopharyngeal carcinoma (NPC) and discriminate it from benign hyperplasia on a non-contrast-enhanced ...MRI sequence for potential use in NPC screening programs.
Methods
We retrospectively analyzed 412 patients who underwent T2-weighted MRI, 203 of whom had biopsy-proven primary NPC confined to the nasopharynx (stage T1) and 209 had benign hyperplasia without NPC. Thirteen patients were sampled randomly to monitor the training process. We applied the Residual Attention Network architecture, adapted for three-dimensional MR images, and incorporated a slice-attention mechanism, to produce a CNN score of 0–1 for NPC probability. Threefold cross-validation was performed in 399 patients. CNN scores between the NPC and benign hyperplasia groups were compared using Student's
t
test. Receiver operating characteristic with the area under the curve (AUC) was performed to identify the optimal CNN score threshold.
Results
In each fold, significant differences were observed in the CNN scores between the NPC and benign hyperplasia groups (
p
< .01). The AUCs ranged from 0.95 to 0.97 with no significant differences between the folds (
p
= .35 to .92). The combined AUC from all three folds (
n
= 399) was 0.96, with an optimal CNN score threshold of > 0.71, producing a sensitivity, specificity, and accuracy of 92.4%, 90.6%, and 91.5%, respectively, for NPC detection.
Conclusion
Our CNN method applied to T2-weighted MRI could discriminate between malignant and benign tissues in the nasopharynx, suggesting that it as a promising approach for the automated detection of early-stage NPC.
Key Points
• The convolutional neural network (CNN)–based algorithm could automatically discriminate between malignant and benign diseases using T2-weighted fat-suppressed MR images.
• The CNN-based algorithm had an accuracy of 91.5% with an area under the receiver operator characteristic curve of 0.96 for discriminating early-stage T1 nasopharyngeal carcinoma from benign hyperplasia.
• The CNN-based algorithm had a sensitivity of 92.4% and specificity of 90.6% for detecting early-stage nasopharyngeal carcinoma.
Reactivation of hepatitis B virus (HBV) infection is a well-recognized complication in cancer patients with chronic HBV (hepatitis B surface antigen HBsAg positive) undergoing cytotoxic chemotherapy. ...In patients who have resolved HBV (HBsAg negative and antibody to hepatitis B core antigen anti-HBc +/- antibody to hepatitis B surface antigen anti-HBs positive), such incidence has been much less common until recent use of rituximab. In this study on HBsAg-negative/anti-HBc-positive lymphoma patients, the objectives were to determine the HBV reactivation rate in patients treated with rituximab-containing chemotherapy and to compare it with the rate in patients treated without rituximab.
Between January 2003 and December 2006, all patients diagnosed with CD20(+) diffuse large B-cell lymphoma (DLBCL) had HBsAg determined before anticancer therapy. They were treated with either cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) alone or rituximab plus CHOP (R-CHOP). HBsAg-negative patients had anti-HBc determined; serum was stored for anti-HBs and HBV DNA. All patients were observed for HBV reactivation, which was defined as detectable HBV DNA with ALT elevation during and for 6 months after anticancer therapy.
Among 104 CD20(+) DLBCL patients, 80 were HBsAg negative. Of the latter, 46 patients (44.2%) were HBsAg negative/anti-HBc positive; 25 of these patients were treated with CHOP, and none had HBV reactivation. In contrast, among the 21 patients treated with R-CHOP, five developed HBV reactivation, including one patient who died of hepatic failure (P = .0148). Exploratory analysis identified male sex, absence of anti-HBs, and use of rituximab to be predictive of HBV reactivation.
Among HBsAg-negative/anti-HBc-positive DLBCL patients treated with R-CHOP, 25% developed HBV reactivation. Close monitoring until at least 6 months after anticancer therapy is required, with an alternative approach of prophylactic antiviral therapy to prevent this potentially fatal condition.
•Lung cancer patients using target therapy frequently progressed in original sites.•Pre-emptive local therapy to residuals after target therapy may defer progression.•ATOM is a phase 2 proof of ...concept study suggesting such strategy is feasible.•The one-year progression free survival since local therapy reached 68.8 %.•Pre-emptive radiotherapy appears to be safe in deferring progression.
NSCLC patients harboring EGFR mutation invariably developed resistance to EGFR TKI. We postulated that oligoresidual disease (ORD) after initial TKI might harbor resistant clones. This study aimed to test if preemptive local ablative therapy (LAT) can improve progression free survival (PFS) or not compared to historic data.
Patients indicated for EGFR TKI who possessed ORD (≤ 4 PET-avid lesions) after an initial 3-month TKI therapy were enrolled. After screening PET-CT, eligible patients with PET-avid ORDs were treated by LAT, either by stereotactic ablative radiotherapy (SABR) or surgery per clinicians’ discretion. TKI was continued after LAT until it was considered ineffective. PET-CT was repeated on the 3rd and 12th month post-LAT (or at progression) apart from regular imaging. Further LAT was allowed in oligoprogressive disease. Primary endpoint was PFS rate at one-year from enrollment. Overall survival (OS), PFS and treatment safety were secondary endpoints. A post hoc comparison with screen failure cohort was performed.
Eighteen patients were enrolled from 2014-17. Recruitment was stopped before the planned number (34) due to slow accrual. Two were excluded due to consent withdrawal and significant protocol violation. Median follow up was 39.1 months. Among the 16 analyzed patients, the one-year PFS rate (i.e. 15 month post TKI) was 68.8 %. Median OS was 43.3 months. All LAT were done by SABR, and none experienced ≥ grade 3 SABR related toxicities. Compared with screen failure cohort (n = 48), pre-emptive LAT effectively reduced risk of progression (HR 0.41, p = 0.0097).
Preemptive LAT in ORD appeared to be safe and feasible. The 1-year PFS rate was encouraging. However, potential biases and the limitations of the study should not be overlooked. Further randomized studies are warranted.
Epigenetic aberrations have been reported in hepatocellular carcinoma (HCC). In this study of patients with unresectable HCC and chronic liver disease, epigenetic therapy with the histone deacetylase ...inhibitor belinostat was assessed. The objectives were to determine dose-limiting toxicity and maximum-tolerated dose (MTD), to assess pharmacokinetics in phase I, and to assess activity of and explore potential biomarkers for response in phase II.
Major eligibility criteria included histologically confirmed unresectable HCC, European Cooperative Oncology Group performance score ≤ 2, and adequate organ function. Phase I consisted of 18 patients; belinostat was given intravenously once per day on days 1 to 5 every 3 weeks; dose levels were 600 mg/m(2) per day (level 1), 900 mg/m(2) per day (level 2), 1,200 mg/m(2) per day (level 3), and 1,400 mg/m(2) per day (level 4). Phase II consisted of 42 patients. The primary end point was progression-free survival (PFS), and the main secondary end points were response according to Response Evaluation Criteria in Solid Tumors (RECIST) and overall survival (OS). Exploratory analysis was conducted on pretreatment tumor tissues to determine whether HR23B expression is a potential biomarker for response.
Belinostat pharmacokinetics were linear from 600 to 1,400 mg/m(2) without significant accumulation. The MTD was not reached at the maximum dose administered. Dose level 4 was used in phase II. The median number of cycles was two (range, one to 12). The partial response (PR) and stable disease (SD) rates were 2.4% and 45.2%, respectively. The median PFS and OS were 2.64 and 6.60 months, respectively. Exploratory analysis revealed that disease stabilization rate (complete response plus PR plus SD) in tumors having high and low HR23B histoscores were 58% and 14%, respectively (P = .036).
Epigenetic therapy with belinostat demonstrates tumor stabilization and is generally well-tolerated. HR23B expression was associated with disease stabilization.
In this prospective cross-sectional study on young premenopausal breast cancer patients, the objectives were to: determine the incidences of chemotherapy-related amenorrhea (CRA) and menopause (CRM); ...identify associated factors; and assess plasma levels of estradiol (E2) and follicular stimulating hormone (FSH) among patients who developed menopause.
Eligibility criteria include Chinese stage I-III breast cancer patients, premenopausal, age ≤45 at breast cancer diagnosis, having received adjuvant chemotherapy, within 3-10 years after breast cancer diagnosis. Detailed menstrual history prior to and after adjuvant treatment was taken at study entry. Patients' background demographics, tumor characteristics and anti-cancer treatments were collected. The rates of CRA and CRM were determined. Analysis was conducted to identify factors associated with CRM. For postmenopausal patients, levels of E2 and FSH were analyzed.
286 patients were recruited; the median time from breast cancer diagnosis to study entry was 5.0 years. 255 patients (91.1%) developed CRA. Of these, 66.7% regained menstruation. At the time of study entry, 137 (48.9%) had developed CRM, amongst whom 84 were age ≤45. On multivariate analysis, age was the only associated factor. Among patients with CRM, the median FSH was 41.0 IU/L; this was significantly lower in those who were taking tamoxifen compared to those who were not (20.1 vs. 59.7 IU/L, p<0.0001). The E2 level was <40 pmol/L; there was no difference between those who were still on tamoxifen or not.
After adjuvant chemotherapy, the majority of young Chinese breast cancer patients developed CRA; ~50% developed CRM, with 61% at age ≤45. Age at diagnosis is the only factor associated with CRM. FSH level may be affected by tamoxifen intake.
Purpose
Convolutional neural networks (CNNs) show potential for delineating cancers on contrast-enhanced MRI (ce-MRI) but there are clinical scenarios in which administration of contrast is not ...desirable. We investigated performance of the CNN for delineating primary nasopharyngeal carcinoma (NPC) on non-contrast-enhanced images and compared the performance to that on ce-MRI.
Materials and methods
We retrospectively analyzed primary NPC in 195 patients using a well-established CNN, U-Net, for tumor delineation on the non-contrast-enhanced fat-suppressed (fs)-T2W, ce-T1W and ce-fs-T1W images. The CNN-derived delineations were compared to manual delineations to obtain Dice similarity coefficient (DSC) and average surface distance (ASD). The DSC and ASD on fs-T2W were compared to those on ce-MRI. Primary tumor volumes (PTVs) of CNN-derived delineations were compared to that of manual delineations.
Results
The CNN for NPC delineation on fs-T2W images showed similar DSC (0.71 ± 0.09) and ASD (0.21 ± 0.48 cm) to those on ce-T1W images (0.71 ± 0.09 and 0.17 ± 0.19 cm, respectively) (
p
> 0.05), and lower DSC but similar ASD to ce-fs-T1W images (0.73 ± 0.09,
p
< 0.001; and 0.17 ± 0.20 cm,
p
> 0.05). The CNN overestimated PTVs on all sequences (
p
< 0.001).
Conclusion
The CNN showed promise for NPC delineation on fs-T2W images in cases where it is desirable to avoid contrast agent injection. The CNN overestimated PTVs on all sequences.
Objective
To investigate the role of diffusion-weighted imaging (DWI) in predicting and monitoring chemoradiotherapy response in head and neck squamous cell carcinoma (HNSCC).
Methods
...Diffusion-weighted imaging was performed pre-treatment (
n
= 50), intra-treatment (
n
= 41) and post-treatment (
n
= 20). Apparent diffusion coefficient (ADC) values were correlated with locoregional failure (LF).
Results
Locoregional failure occurred in 20/50 (40%) patients. A significant correlation was found between LF and post-treatment ADC (
p
= 0.02) but not pre- or intra-treatment ADC. Serial change in ADC was even more significant (
p
= 0.00001), using a fall in ADC early (pre- to intra-treatment) or late (intra- to post-treatment) to indicate LF, achieved 100% specificity, 80% sensitivity and 90% accuracy.
Conclusions
Single ADC measurements pre- or intra-treatment did not predict response, but ADC post-treatment was a marker for LF. Serial change in ADC was an even stronger marker, when using an early or late treatment fall in ADC to identify LF.
The oncogenic PI3K/Akt/mTOR pathway is frequently activated in HCC. Data on the mTOR inhibitor, temsirolimus, is limited in HCC patients with concomitant chronic liver disease. The objectives of this ...study were: (1) In phase I, to determine DLTs and MTD of temsirolimus in HCC patients with chronic liver disease; (2) In phase II, to assess activity of temsirolimus in HCC, and (3) to explore potential biomarkers for response.
Major eligibility criteria included histologically confirmed advanced HCC and adequate organ function. In Phase I part of the study, temsirolimus was given weekly in 3-weekly cycle; dose levels were 20 mg (level 1), 25 mg (level 2) and 30 mg (level 3). The MTD was used in the subsequent phase II part; the primary endpoint was PFS and secondary endpoints were response and OS. In addition, exploratory analysis was conducted on pre-treatment tumour tissues to determine stathmin, pS6, pMTOR or p-AKT expressions as potential biomarkers for response. Overall survival and PFS were calculated using the Kaplan-Meier method. Reassessment CT scans were done every 6 weeks. All adverse events were reported using CTCAE v3.
The Phase I part consisted of 19 patients, 2 of 6 patients at level 3 experienced DLT; dose level 2 was determined to be the MTD. The phase II part consisted of 36 patients. Amongst 35 assessable patients, there were 1 PR, 20 SD and 14 PD. Overall, the median PFS was 2.83 months (95% C.I. 1.63-5.24). The median OS was 8.89 months (95% C.I. 5.89-13.30). Grade ≥ 3 that occurred in > 10% of patients included thrombocytopenia (4) and hyponatraemia (4). Exploratory analysis revealed that disease stabilization (defined as CR + PR + SD > 12 weeks) in tumours having high and low pMTOR H-scores to be 70% and 29% respectively (OR 5.667, 95% CI 1.129-28.454, p = 0.035).
In HCC patients with chronic liver disease, the MTD of temsirolimus was 25 mg weekly in a 3-week cycle. The targeted PFS endpoint was not reached. However, further studies to identify appropriate patient subgroup are warranted.
This study has been registered in ClinicalTrials.gov (Id: NCT00321594) on 1 December 2010.
There are limitations in using radiologic evaluation to assess the treatment outcome of patients with hepatocellular carcinoma (HCC). The use of serial alpha-fetoprotein (AFP) in monitoring response ...has not been rigorously evaluated. We aimed to study the clinical value of AFP trend in an attempt to validate AFP as a surrogate serologic end point.
Participants from a phase III randomized trial of systemic chemotherapy in HCC were studied. Serum AFP was prospectively collected in parallel with clinical and radiologic outcome. AFP response was defined as a decrease in AFP of more than 20% after a minimum of two cycles of chemotherapy. We studied the relationship between AFP response and treatment outcome in terms of radiologic response and overall survival.
Of 188 patients, 117 patients with elevated serum AFP (> 20 microg/L) and documented radiologic evaluation had received at least two cycles of chemotherapy. A total of 47 AFP responders were identified. AFP responders had better survival than nonresponders (13.5 v 5.6 months, respectively; P < .0001), and AFP response was strongly associated with radiologic response (P < .0001). Multivariate analysis suggested that AFP response was significantly associated with survival (hazard ratio, 0.413; 95% CI, 0.273 to 0.626; P < .0001). AFP responses were frequently observed in patients with radiologically stable disease (SD) and tended to identify a subgroup of SD patients with better survival.
Serial AFP measurement is useful in prognostication and monitoring treatment response in HCC patients undergoing systemic chemotherapy. Incorporation of AFP response into the criteria evaluating treatment outcome should be considered in clinical practice and clinical trials of novel agents in HCC.