Keywords: cancer suspicion
, health literacy
, salvage androgen deprivation therapy
, breast cancer
, assisted reproductive technology
, pancreatic cancer
, breast density
Selected reports edited by Janet Fricker
Higher quality colonoscopies reduce cancer deaths without higher costs
Higher rates of detection of precancerous adenoma were associated with lower lifetime risks of colorectal cancer and colorectal cancer mortality and did not result in greater overall healthcare costs, a Dutch microsimulation model study has found.
Colonoscopy reduces colorectal cancer mortality through detection and treatment of precursor adenomatous or early cancerous lesions. However, quality, as measured by adenoma detection rates, varies widely among physicians. While studies suggest that higher adenoma detection rates are associated with better disease detection and better management, little is known about consequences for costs and other benefits of screening programmes.
Reinier Meester and colleagues, from Erasmus University Medical Centre, Rotterdam, carried out microsimulation modelling to estimate the lifetime benefits, complications and costs of an initial colonoscopy screening programme at different levels of adenoma detection. The team used data from the Kaiser Permanente Northern California healthcare system on variations in adenoma detection rates and cancer for 57,588 patients examined by 136 gastroenterologists between January 1998 and December 2010.
For the study, no screening was compared with colonoscopy screening according to adenoma detection rate quintiles (divided into five groups). The mean adenoma detection rates were 15.32% for quintile 1, 21.27% for quintile 2, 25.61% for quintile 3, 30.89% for quintile 4 and 38.66% for quintile 5.
The model estimated that, among unscreened patients, lifetime colorectal cancer risk was 34.2 cases per 1,000 patients, compared to 26.6 per 1,000 for those in quintile 1, 21.6 per 1,000 for those in quintile 2, 19.0 per 1,000 for quintile 3, 15.6 per 1,000 for quintile 4, and 12.5 per 1,000 for quintile 5. The simulated incidence of death from colorectal cancer was 13.4 per 1,000 patients for unscreened patients compared to 5.7 per 1,000 for those in quintile 1, 4.5 per 1,000 for those in quintile 2, 3.7 per 1,000 for those in quintile 3, 3.0 per 1,000 for quintile 4 and 2.3 per 1,000 for quintile 5.The model estimated that lifetime incidence and mortality risks were 11%–13% lower on average for every five-point higher adenoma detection rates, which translates to overall differences of 53–60% between the lowest and highest quintiles. Estimated net screening costs were on average 3.2% lower for every five-point increase in adenoma detection rates, and the risk of complications was on average 9.8% higher for every five-point increase in adenoma detection rates.
“Our results suggest that higher adenoma detection rates may be associated with up to 50%–60% lower lifetime colorectal cancer incidence and mortality without higher net screening costs, despite a higher number of colonoscopies and polypectomy-associated complications,” write the authors. Future research, they add, is needed to assess why adenoma detection rates vary and whether increasing adenoma detection would be associated with improved patient outcomes.
- R Meester, C Doubeni, I Lansdorp-Vogelaar et al. Variation in adenoma detection rate and the lifetime benefits and cost of colorectal cancer screening. A microsimulation model.JAMA 16 June 2015, 313:2349–58
Very low breast density predicts worse survival
=> European Radiology
Very low mammographic breast density (MBD) at the time of breast cancer diagnosis is associated with higher tumour grade and predicts poorer disease-free and overall survival, a Finnish study has found.
MBD refers to the appearance of breast tissue on mammograms, reflecting variations in breast tissue composition and x-ray attenuation characteristics. Unlike other breast cancer risk factors, MBD – which is influenced by genetic factors – can change over time. It decreases with age and is further reduced by multiparity and menopause. While it is well recognised that MBD in the upper quartile is associated with a four- to six-fold higher risk for developing breast cancer than MBD in the lower quartile, less is known about its possible prognostic importance.
Since high MBD is such a strong risk factor, it could be hypothesised that, in patients with breast cancer, higher densities would yield a worse prognosis. In the current study Amro Masarwah and colleagues, from Kuopio University Hospital, Finland, set out to examine the prognostic value of MBD and other mammographic features in 270 patients, who had a median age of 58 years, with previously diagnosed invasive breast cancer. MBD was classified by consensus among five trained radiologists according to density, with very low density (VLD) <10%; low density (LOD) <25% and mixed density (MID) >25%.
Results at a mean follow-up of 6.4 years showed that disease-free survival was 74.7% (118/158) for patients with LOD versus 84.8% (95/112)for patients with MID (P=0.048), and that overall survival was 75.3% for patients with LOD versus 90.2% for patients with MID (P=0.003). In Cox regression analysis, in comparison to the other groups VLD proved to be an independent feature predicting poor prognosis (HR=3.275; P<0.001) that was second in importance only to tumour size (HR=3.455; 95% CI 1.833–6.511; P<0.001). Percentile MBD categories displayed a significant inverse relationship with tumour grade (P=0.019), but had no relation to HER2 status, or oestrogen or progesterone receptor status.
“Breast density is a readily available, cheap and easy-to-interpret form of information and, according to our analysis, proved to be an independent and clinically important prognostic feature. The ability to predict the course and outcome of the disease beforehand by analysing certain features on a mammogram is a desirable and useful tool for clinicians,” write the authors.
In future breast cancer studies, they add, proper categorisation of breast tissue density is important. “Performing a more detailed radiological subdivision amongst the low density group should be advocated, as only the patients with very low densities showed significant associations with poor survival,” write the authors. Studies are needed to clarify hormonal, biological and genetic interconnections between breast density and breast cancer aggressiveness.
- A Masarwah, P Auvinen, M Sudah et al. Very low mammographic breast density predicts poorer outcome in patients with invasive breast cancer. Eur Radiol July 2015, 25: 1875–82
Biomarkers identified for screening pancreatic cancer
=> Clinical Cancer Research
A three-protein biomarker panel screening test undertaken in urine can be used to detect patients with early-stage pancreatic cancer, report UK researchers.
Despite progress in understanding pancreatic ductal adenocarcinoma (PDAC) at the molecular level, no significant improvements in diagnosis and therapy have been made in the last 30 years. Currently 80% of patients with PDAC present with locally invasive and/or metastatic disease, leaving only 20% eligible for potentially curative surgery. Timely detection of PDAC has been hampered by lack of specific clinical symptoms in the early stage of disease, insufficient sensitivity of current imaging modalities, and lack of accurate body fluid-based biomarkers for early stage disease.
In the current study Tatjana Crnogorac-Jurcevic and colleagues, from Queen Mary University, London, looked to develop a diagnostic test in urine specimens. Urine, the investigators reasoned, offers advantages over blood since it provides an ‘inert’ and stable matrix for analysis and can be repeatedly and noninvasively sampled in sufficient volumes.
In the discovery phase the team explored levels of 1,500 proteins in urine specimens from six patients with PDAC, six patients with chronic pancreatitis (CP) and six healthy controls. Using GELC/MS/MS assays they found that only 481 of the proteins were common to males and females, and that three of these proteins – LYVE-1, REG1A and TFF1 – were higher in PDAC patients.
For the validation phase, 192 urine samples from PDAC patients, 92 from chronic pancreatitis patients and 87 from healthy subjects were assayed.
When comparing PDAC specimens with healthy urine specimens, the resulting areas under the receiver-operating characteristic curves (which plot true-positives against false-positives, providing information on detection accuracy) were 0.89 (95%CI 0.84–0.94) in the training dataset and 0.92 (95%CI 0.86–0.98) in the validation dataset. When comparing PDAC stage I–II (n=71) with healthy urine specimens, the panel achieved areas under the curves of 0.90 (95%CI 0.84–0.96) in the training dataset and 0.93 (95%CI 0.84–1.00) in the validation dataset. Furthermore, an exploratory analysis suggested accuracy was increased when the panel was combined with the CA19.9 protein for patients with PDAC, but not those with stage I-IIA PDAC.
“Being completely non-invasive and inexpensive, this urine screening test could, upon further validation, and when coupled with timely surgical intervention, lead to a much improved outcome in patients with high risk of developing PDAC,” write the authors.
The healthy cancer controls, they add, were on average younger than cancer patients, making it important for further validation studies to use older controls.
The improved accuracy with CA19.9, say the authors, may be important in light of recent findings that serum CA19.9 is upregulated for up to two years before PDAC diagnosis.
The priority cohort for screening strategies, they suggest, should include families with a high incidence of pancreatic cancer (at least two affected first-degree relatives) and individuals with hereditary conditions, such as intestinal polyposis syndrome Peutz-Jeghers.
- T Radon, N Massat, R Jones et al. Identification of a three-biomarker panel in urine for early detection of pancreatic adenocarcinoma. Clin Cancer Res 1 August 2015, 21:3512–21
Assisted reproductive technology has no influence on breast cancer outcomes
=> European Journal of Cancer
Pregnancy using assisted reproductive technology (ART) in women with a history of breast cancer is feasible and does not appear detrimental to cancer outcomes, a European multicentre retrospective study has found.
With around 65–70% of young breast cancer patients alive and free of distant relapse 10 years after diagnosis, there is a need to explore the feasibility and safety of ART in breast cancer survivors whose fertility may have been impaired by treatment.
In the current study Oranite Goldrat, from Erasme Hospital in Brussels, Belgium, and colleagues from Brussels, Milan, Macerata, Barcelona and Denmark, for the first time set out to evaluate the effect of using assisted reproductive technology on rates of recurrence and death in patients who were previously treated for breast cancer and subsequently became pregnant.
Women aged 18 to 45 years, who were diagnosed with primary non-metastatic breast cancer between 2000 and 2009 and who subsequently became pregnant, were eligible for the study. The cohort was divided into two groups according to whether pregnancies occurred spontaneously (Spontaneous Group) or after using assisted reproductive technology (ART Group). Data were collected on clinico-pathological characteristics, breast cancer treatment (date of diagnosis, histological type, histological grade, tumour size, nodal status, endocrine receptor status, HER2 status, type of breast surgery, chemo- and endocrine therapies), fertility treatments (ovulation induction, ovarian stimulation for IVF and oocyte donation) and pregnancy-related information (age at conception, number of pregnancies, and pregnancy outcome).
Results showed that altogether 173 women were followed up in the Spontaneous Group (247 pregnancies) and 25 women in the ART group (34 pregnancies). No significant differences in breast cancer outcomes were observed between the two groups for local recurrence, distant recurrence and contralateral breast cancer (P=0.54 for all). Patients in the spontaneous pregnancy group were younger (mean age: 31.2 vs 33.7 years, P=0.009) and had a higher frequency of histological grade 3 tumours (59.6% vs 36%, P=0.033). On the other hand, women in the ART group had more node-negative, oestrogen receptor- (ER)-positive tumours and shorter durations of endocrine therapy, although these differences did not reach statistical significance.
“Our results indicate lack of a detrimental effect of attaining pregnancy via ART on the risk of recurrence in women with history of breast cancer. While the number of patients included in the study is relatively small, warranting further confirmation, we believe this study would provide physicians with important guidance when counselling their patients in the daily practice,” write the authors.
Interestingly, they add, women undergoing ART had more favourable prognostic parameters, suggesting physicians were selective in offering ART to patients with a relatively good prognosis. “This underscores the uncertainty and fear of the safety of ART in women with history of breast cancer,” write the authors.
The trend for earlier discontinuation of endocrine therapy among ART patients, they suggest, is due to higher age, leaving women with no choice but to discontinue treatment early.
- O Goldrat, N Kroman, F Peccatori et al. Pregnancy following breast cancer using assisted reproduction and its effect on long-term outcome. Eur J Cancer August 2015, 51:1490–96
Anxiety and health literacy are drivers for salvage androgen deprivation therapy
=> Annals of Oncology
Among men with asymptomatic prostate-specific antigen (PSA) recurrence after radiotherapy for prostate cancer, use of salvage androgen deprivation therapy was nearly twice as high among men with high levels of anxiety about their PSA or poor health literacy, a US multicentre prospective registry has found.
Although androgen deprivation therapy is part of the standard approach for the initial management of metastatic prostate cancer, no survival benefits have been shown from early use of salvage androgen deprivation therapy for men with PSA-only recurrence after radiotherapy (who may not go on to develop overtly metastatic disease for years).
In the study, Paul Nguyen and colleagues, from Dana Farber Cancer Institute and Brigham and Women’s Hospital, used the prospective Comprehensive, Observational, Multicenter, Prostate Adenocarcinoma Registry (COMPARE) to determine whether PSA anxiety or health literacy are associated with unproven use of early salvage androgen deprivation therapy as initial management for PSA recurrence following radiotherapy.
Between February 2004 and March 2007, the COMPARE registry enrolled 1,120 men across 150 geographically diverse sites with biochemical (PSA) recurrence after primary therapy for localised prostate cancer. A total of 375 participants met the inclusion criteria of having received radiotherapy (external beam radiotherapy or brachytherapy) but not radical prostatectomy as initial treatment, and having complete information on PSA anxiety (assessed using three questions from the Memorial Anxiety Scale for Prostate Cancer) and health literacy (based on the Rapid Estimation of Literacy of Medicine [REALM-SF]). For REALM-SF, patients were asked to pronounce seven medical words out loud (meno-pause, exercise, rectal, behaviour, antibiotics, jaundice, and anaemia), to give literacy levels higher than ninth grade (all words pronounced correctly) and lower than ninth grade (fewer than seven words pronounced correctly).
Results showed 68 men (18.1%) received salvage androgen deprivation therapy as initial management for PSA recurrence. For men with high PSA anxiety, 28.8% received salvage androgen deprivation therapy compared to 13.1% who did not have high anxiety (univariable OR=2.15, 95%CI 1.16–4.00, P=0.0015; multivariable OR 2.36; 95%CI 1.21–4.62; P=0.012). For men who had higher level of health literacy, 15.2% underwent salvage androgen deprivation therapy compared to 26.3% with lower levels of health literacy (univariable OR=0.50; 95%CI 0.29–0.88, P=0.016; multivariable OR=0.58; 95%CI 0.32–1.05; P=0.07).
“These findings suggest that patient-level concern or understanding are significant drivers… of receipt of salvage ADT [androgen deprivation therapy] at biochemical recurrence after radiotherapy and men with high PSA anxiety or low health literacy may be more likely to push for or accept treatment,” write the authors.
Given that early salvage androgen deprivation therapy is costly, worsens quality of life, and has not been shown to improve survival, write the authors, quality improvement strategies are needed to help these individuals. “Oncologists should ensure that anxious and less health literate patients are adequately counselled about the benefits and risks of salvage ADT before they decide to pursue therapy. A concerted effort to ensure that patients’ worries, anxieties, and uncertainties are addressed could prevent many patients from undergoing early initiation of salvage ADT.”
Educational attainment influences suspicion of cancer
- B Mahal, M Chen, C Bennett et al. High PSA anxiety and low health literacy skills: drivers of early use of salvage ADT among men with biochemically recurrent prostate cancer after radiotherapy. Annal Oncol July 2015, 26:1390–95
=>European Journal of Cancer
Levels of cancer suspicion following ‘warning symptoms’ were low overall, and even lower among less educated respondents, a community sample questionnaire has found.
Studies where cancer patients report retrospectively on the process of symptom appraisal have suggested that failing to recognise symptoms that are due to cancer is an important factor determining delays in presentation. Prolonged intervals from symptom onset to seeking help may increase the risk of being diagnosed at a late stage. It has also been found that people with lower socioeconomic status are more likely to be diagnosed with later-stage disease for several cancer sites.
For the current study, Katriina Whitaker and colleagues, from the University of Surrey, in Guildford, UK, emailed two separate primary-care-based symptom surveys (using the same questions) to 9,771 adults aged over 50 years, with no cancer diagnosis, to test the hypothesis that people with less education are less likely to suspect cancer when they experience a cancer ‘warning sign’.
Respondents were asked whether they had experienced any of 10 cancer ‘warning signs’, taken from the Cancer Research UK’s website, in the past three months. The warning signs were: persistent cough or hoarseness, persistent change in bowel habits, persistent unexplained pain, persistent change in bladder habits, change in appearance of a mole, unexplained lump, sore that does not heal, unexplained weight loss, persistent difficulty swallowing and unexplained bleeding. All had yes/no responses, and for each symptom respondents experienced they were asked: “What do you think caused it?” Surveys also included questions about marital status, employment, ethnicity, and education (university versus below university).
Results showed that nearly half the respondents (1,790/3,756) had experienced a ‘warning sign’, but only 3.5% (63/1,790) of these mentioned cancer as a possible cause. The highest number of cancer suspicions was for change in the appearance of a mole (10.7%), while the lowest number was for a change in bladder habits (0.7%).
Lower education level was associated with lower likelihood of cancer suspicion: 2.6% of respondents with school-only education versus 7.3% with university education suspected cancer as a possible cause. In multivariable analysis, low educational level was the only demographic variable independently associated with lower cancer suspicion (OR=0.34). There were no significant associations with sex, age, marital status, employment or ethnicity.
“Our finding that people in general have low cancer suspicion when they experience ‘warning signs’, and that this is even lower in those more likely to be diagnosed at a later stage is important. People may need to be encouraged to lower their cancer suspicion ‘threshold’ through earlier diagnosis interventions, both at the public health and GP level,” write the authors.
One issue for consideration, they add, is the tension between encouraging people to think seriously about symptoms that could give an early warning of cancer and creating fear or hypochondria. A possible solution, they suggest, is to develop educational information that associates symptoms with potential illness rather than cancer.
- K Whitaker, K Winstanley, U Macleod et al. Low cancer suspicion following experience of a cancer ‘warning sign’. Eur J Cancer (in press) doi.org/10.1016/j.ejca.2015.07.014