Showing papers by "Talha Khan Burki published in 2016"

Sugar tax in the UK

Abstract: www.thelancet.com/oncology Vol 17 May 2016 e182 UK Chancellor of the Exchequer, George Osborne, made a surprise announcement during last week’s budget. From 2018, drinks in which the sugar content exceeds 8 grams per 100 millilitres will be liable to a levy, equivalent to 24 pence per litre. As things stand, this will aff ect the regular versions of Coca-Cola, Pepsi Cola, and Red Bull. Drinks with a sugar content of 5–8 grams per 100 millilitres will be taxed at the equivalent of 18 pence per litre. Fruit juices and milk-based drinks will be exempt (although some of these contain far more sugar than a typical soft drink). The Government expects the tax to raise £520 million in the fi rst year, money which will be spent on school sports to encourage healthier behaviour. The focus on fizzy drinks may be driven by expediency—these are highly visible items which provide no nutritional benefit. A tax on confectionery or processed food would be tricky to formulate. Besides, the main target is British children and teenagers, around one-third of whom are overweight or obese. According to some estimates, every day the average 11–18 year old in the UK consumes roughly three times their recommended quantity of sugar. Soft drinks appear to be this group’s single largest source of sugar. “Overweight and obese children are more likely to be overweight and obese adults, and that increases your risk of up to ten different cancers”, said Chit Selvarajah of Cancer Research UK, London, UK. A report commissioned by the organisation suggested that if current trends continue, the UK will face some 670 000 new cases of cancer caused by obesity over the next 20 years. “It is brilliant to see bold action from the Government on this issue—it is a really important first step in terms of addressing childhood obesity”, Selvarajah told The Lancet Oncology. S imon Capewel l f rom the Faculty of Public Health, (London, UK), expressed qualms over the structure of the levy. “Industry will work very hard to reduce sugar content to 1% less than the cut-off point”, he explained. “A tax graded gram-by-gram would exert steady pressure on them to reduce the sugar content down towards zero”. But he welcomed the measure. “The Government has stated very publicly that they have looked at the evidence of harm from sugar and been persuaded, and it is extremely symbolic, important, and powerful that they have put the levy on drinks manufacturers”, Capewell concluded.

Arginine deprivation for ASS1-deficient mesothelioma.

Abstract: www.thelancet.com/oncology Vol 17 October 2016 e423 Arginine deprivation therapy improves progression-free survival in patients with argininosuccinate synthetase 1 (ASS1)-defi cient malignant pleural mesothelioma, according to a new study . The investigators in the phase 2 trial screened 201 patients with mesothelioma from centres in the UK. 97 (48%) of 201 patients showed ASS1 defi ciency. 68 patients were randomly assigned to receive either the arginine-lowering agent pegylated arginine deiminase (ADI-PEG20; 36·8 mg/m2 per week for up to 6 months) plus best supportive care (n=44), or best supportive care alone (n=24). Median progressionfree survival in the ADI-PEG20 arm was 3·2 months versus 2·0 months in the best supportive care group (hazard ratio [HR] 0·56; 95% CI 0·33–0·96, p=0·03). “In mesothelioma, the longer the duration of arginine deprivation in the treated group, the longer the progression-free survival”, noted lead author Peter Szlosarek (Queen Mary University of London, London, UK). Serious adverse events were similar in both groups. Moreover, the results showed that patients with ASS1 loss exceeding 75% received a greater benefi t from arginine deprivation than those with ASS1 loss of 50–75% (PFS HR 0·25; 95% CI 0·09–0·70; vs HR 0·72; 95% CI 0·34–1·49; interaction p=0·21). “ASS1 is a predictive and prognostic biomarker,” explained Szlosarek. “We are always looking for things that do not have much toxicity, but are also effi cacious”, said Harvey Pass (New York University School of Medicine, NY, USA). “This is a small study, but it is very promising—the arginine pathway looks extremely interesting.” Countries such as India and China continue the heavy use of asbestos. Over the coming decades, the global burden of asbestos-related disease (eg, mesothelioma) is expected to sharply rise. Even the UK, which banned all forms of asbestos in 1999, has yet to reach its epidemic peak. Establishing diagnostic, predictive, and prognostic biomarkers for mesothelioma is a vital task. “That is an important aspect of this paper—the emphasis on combining treatment with a biomarker is where we need to go,” Pass notes. Szlosarek believes that the potential of arginine deprivation as a monotherapy is probably limited to patients with high ASS1 loss. “We see arginine deprivation as something to be used in combination,” said Szlosarek. His group is now studying ADI-PEG20 in tandem with standard chemotherapy for patients with mesothelioma.

Resistance to PD-1 blockade in melanoma

Abstract: www.thelancet.com/oncology Vol 17 September 2016 e376 A new study by Antoni Ribas and colleagues has identified two likely mechanisms for the development of resistance to PD-1 blockade in patients with melanoma. The investigators did whole-exome sequencing on tumour samples taken from four patients with metastatic melanoma who had relapsed after treatment with pembrolizumab. The patients had all shown an objective response to pembrolizumab for at least 6 months and mean time to relapse was 624 days (range 419–888). Biopsies were taken before treatment began and after disease progression. The tumours in two of the patients showed over 90% of the same mutations both before and after treatment. But their relapsed tumours had developed mutations in the JAK1 and JAK2 genes, respectively. “These were both loss-of-function mutations that had become homozygous after losing the wild type copy”, explained co-author Jesse Zaretsky (University of California, Los Angeles, Los Angeles, CA, USA). “The tumours had worked very hard to get rid of those genes.” JAK1 and JAK2 are crucial for mediating interferon-related signalling. The third patient’s relapsed tumour showed a mutation in B2M, the gene for beta-2-microglobulin, which has been previously implicated in immunotherapy resistance. Without this protein, tumour cells no longer present antigens to T cells. So whereas mutations in JAK1 and JAK2 suggest that the tumour is insensitive to T cells inhibition, a mutation in B2M means the T-cells cannot recognise the tumour in the fi rst place. The fourth patient had no apparent mutations related to acquired resistance to T cells. Zaretsky cautioned that the sample size precludes drawing conclusions on the frequency of these mutations in relapsed patients. “It is also possible that other resistance mechanisms or immune suppressive factors could be at play at the same time”, he said. Ramesh Rengan (University of Washington, Seattle, WA, USA) wondered whether the same resistance pathway is activated at all sites of disease within the same patient. “It would be very diffi cult to re-establish the interferon-signalling pathway”, he explained. “But if we bring the patient to a reasonable response from the immunotherapy, if we see isolated progression, we could try a different approach.” If immunotherapy is no longer viable, this approach might be radiotherapy, or a therapy that specifi cally targets the cancer cells.

New risk loci for endometrial cancer identified.

Abstract: www.thelancet.com/oncology Published online May 12, 2016 http://dx.doi.org/10.1016/S1470-2045(16)30135-8 1 New research has identifi ed fi ve gene regions associated with a heightened risk of endometrial cancer. The study consisted of a meta-analysis of three genomewide association studies and two follow-up phases, involving 7737 cases of endometrial cancer and 37 144 controls. The meta-analysis and genome-wide imputation yielded five new risk loci: 13q22.1 (rs11841589, close to KLF5), 6q22.31 (rs13328298, near HEY2 and NCOA7), 8q24.21 (rs4733613, near MYC and rs17232730), 15q15.1 (rs937213, in EIF2AK4) and 14q32.33 (rs2498796, in AKT1, near SIVA1). “Endometrial cancer has really been lagging behind other cancers, with only two known signals, so this is a relatively large step forward, but there is still a long way to go”, explained author Ian Tomlinson (Oxford University, Oxford, UK). The fi nding that most interested his group was the association near KLF5. They performed functional studies of the 13q22.1 locus. “KLF5 is a gene with known potential importance for endometrial cancer—together we pinned down the functional variation that was associated with disease risk”, said Tomlinson. Most of the other genes identifi ed by the investigators had not previously been associated with endometrial cancer, although several have been linked to other cancer types. “In terms of understanding how the disease works, it is important that KLF5 is now fl agged as a target, but we still need a lot more validation”, Tomlinson told The Lancet Oncology. For now, the implications for patients are limited. The fi ndings only account for about 5% of the inherited risk of endometrial cancer. “The main risk factors are age, obesity, and physical activity”, points out Emma Smith (Cancer Research UK, London, UK). “The results from this study might be relevant for women who are at a higher risk of womb cancer, because of a family history. In future, doctors can monitor them for early signs of the disease.” Endometrial cancer prevention is a slightly controversial topic, given the lack of large studies demonstrating the efficacy of any single measure on a population-scale. “Our eventual aim, as for most of the common cancers, is ideally for some sort of chemoprevention”, said Tomlinson. “If we found a battery of these single nucleotide polymorphisms—50 or 60, like there are for prostate or breast cancer—and identified the specific pathways involved, we could start to seriously think about that”.

Counting the cost of end-of-life cancer care

Abstract: www.thelancet.com/oncology Vol 17 March 2016 e91 Two new studies have examined end-of-life care for patients with cancer. Alexi Wright (Harvard University, Boston, MA, USA) and colleagues assessed data for US patients on Medicare who died from lung or colorectal cancer before the end of 2011. They conducted 1146 interviews with family members of these patients. Family members of patients who received hospice care for 3 or fewer days, or not at all, were signifi cantly less likely to rate the end-of-life care as excellent than were family members of patients who received hospice care for more than 3 days. “It is increasingly clear that how and where patients die has a powerful infl uence on patients’ dying experiences, and how family members remember it”, explained Wright. Her study found that 57% of patients died in their preferred location. “Often how a patient wishes to die is not addressed by the medical community”, points out Carol Spence (National Hospice and Palliative Care Organization, Alexandra, VA, USA). Wright agrees. “The system is somewhat biased towards intensive medical care—the only way for patients to receive the care that they want is to articulate it; the key is early conversations”, she told The Lancet Oncology. The second study retrospectively compared differences in end-of-life care for patients with cancer who died in 2010 in several Western European and North American nations. Patients in the USA were least likely to die in hospital: 22% compared with 52% in Canada and 42% in England. The well-established US hospice system helps explain this disparity; the country has more than 5300 hospices, most of which provide home care. But American patients were more than twice as likely to be admitted to the intensive care unit (ICU) within 180 days of death than were patients in any of the six comparator countries. Lead author Ezekiel Emanuel (University of Pennsylvania, Philadelphia, PA, USA) puts this down to the USA’s large supply of ICU beds. “Where other countries put patients into regular hospital beds or keep them at home, we admit them to the ICU”, he said. Mean per capita hospital expenditure for care during the last 180 days of life ranged from US $9342 in England to $21 840 in Canada. “All countries can improve end-of-life care”, noted Emanuel. “Policies need to encourage better attention to patients, better attention to symptoms, and less hospitalisation—the USA especially needs to fi gure out how to stop using the ICU so much”.

Attitudes and practices towards legal euthanasia.

Abstract: www.thelancet.com/oncology Vol 17 August 2016 e325 Findings from a review have suggested that in countries and jurisdictions that permit euthanasia or physicianassisted suicide, the practices are infrequent and largely restricted to patients with cancer. Both practices are legal in Belgium, Canada (since June, 2016), Luxembourg, and the Netherlands. Five US states have legalised assisted suicide, most recently California (the others are Montana, Oregon, Vermont, and Washington). “In western Europe, support for euthanasia and assisted suicide seems to be steadily increasing”, notes lead author Ezekiel Emanuel (University of Pennsylvania, PA, USA). Across the territories where the interventions are legal, between 0·3% and 4·6% of all deaths were reported as assisted suicide or euthanasia. Around 75% of individuals who choose to end their lives by these means have cancer. Pain does not tend to be the main motivating factor; instead patients commonly cited loss of autonomy and dignity, no longer enjoying life, and depression as their reasons for wanting to die. No more than 5% of these patients in Oregon and Washington received psychiatric assessment. “Depression and psychological distress tend to be the signifi cant drivers for euthanasia and assisted suicide, and yet we are not involving psychiatrists, psychologists, and social workers”, said Emanuel. “We ought to look closely at the safeguards we have in place around mental health issues”. Juliet Gichon (University of Calgary, AB, Canada) agrees. “The principle of assisted dying is underpinned by autonomy”, she explained. “That requires patients to be competent, well-informed, and not acting under coercion”. If an individual is depressed, they might not be competent to take such a serious decision. The review also suggested that some patients might have complications when attempting to end their lives. Complications seem to be higher with assisted suicide. “There were clear rates of patients waking up, not going into coma, and vomiting up medication”, Emanuel said. Some of these complications might be attributable to the laws themselves—assisted suicide can be problematic for patients who have difficulty ingesting, therefore in countries that permit euthanasia, a lethal injection can be applied instead. But as with so much in this area, data are scarce. “The amount of research on euthanasia and assisted suicide has gone down substantially, even as more places are legalising the practices”, said Emanuel. “We need a lot more rigorous, up-to-date research”.

Association between allergies and reduced risk of glioma.

Abstract: Authors of a new study have noted an inverse association between the risk of developing glioma and a history of allergies. The researchers drew data from the Glioma International Case Control Study, which contains detailed information about atopic disorders for 4533 cases and 4171 controls in 14 sites and fi ve countries. They concluded that a history of respiratory allergies reduces the risk of developing glioma (meta-analysis odds ratio 0·72 [95% CI 0·58–0·90]), as does having asthma (OR 0·77 [0·64–0·93]) and eczema (OR 0·71 [0·56–0·89]), compared with patients without allergies. “Our current study adds substantially to the weight of the already existing evidence for a true relationship between allergies and glioma”, explained lead author Melissa Bondy (Baylor College of Medicine, Houston, TX, USA). But she cautioned that until the underlying biological mechanism is identified, definitive confirmation of this relationship will not be possible. One possibility is that the heightened state of immunosurveillance that prompts allergic reactions forestalls development of tumours. “More specifically, it has been proposed that IgE antibodies, which are involved in allergic reactions, may have some cross-reactivity to brain tumours, quickly triggering an immune response against new or forming tumours”, said Bondy. She posits that individuals who have a strong allergic response might fi nd it easier to expel toxins and carcinogens from their systems. “It is very difficult to know at what level the interaction is working; it may be something very fundamental in people’s immune systems that makes the difference; we do not have any idea at the moment”, points out Susan Short (Leeds University, Leeds, UK). She places the findings in the context of the increasing recognition that the immune environment strongly affects tumour response in patients with gliomas. “So it is not very surprising that it may make a difference to tumour initiation”, she said. For now, clinical practice is unlikely to be affected—doctors can hardly advise patients to develop protective allergies. “But if we can determine which aspect of the allergic response may be protective against glioma, we might be able to take advantage of that aspect of the immune response to help prevent or treat glioma in the future”, Bondy added.

Defining the genetics of melanoma progression

Abstract: A new study has mapped the genetic progression from precursor lesion to melanoma. The researchers sequenced 293 cancer-relevant genes in 150 areas of 37 primary melanomas and their adjacent precursor lesions. “In every case, a mutation known to activate the mitogen-activated protein kinase signaling pathway, usually in [melanoma associated oncogenes] BRAF or NRAS, could be nominated as the putative initiating oncogene”, they noted. Benign lesions exclusively contained BRAF mutations. Intermediate lesions showed NRAS mutations and other driver mutations. TERT promoter mutations were common in both intermediate lesions and melanomas in situ. PTEN and TP53 mutations were characteristic of advanced tumours, whereas biallelic inactivation of CDKN2A occurred in only invasive melanoma. The authors also noted “a strong signature of the eff ects of UV radiation detectable at all evolutionary stages”. Therefore, UV radiation both triggers melanoma and accelerates its progress. “It means we need to refine our prevention messages”, explains lead author Boris Bastian (University of California, San Francisco, CA, USA), adding that “we need to focus extra attention on protecting pre-existing moles”. He pointed out that their results showed that diff erent types of melanoma had diff erent evolutionary trajectories. BRAF mutations implied that the melanoma had arisen from a benign lesion. “These nevi are already invasive, so to speak—our data suggests that the transformation is happening on the surface of the lesion, where the cells are continually exposed to UV radiation”, said Bastian. Conversely, NRAS mutations or BRAF and BRAF mutations implied that the melanoma emerged from intermediate lesions or a melanoma in situ. “This is a very important piece of work”, Victoria Sanz-Moreno (King’s College London, London, UK) told The Lancet Oncology. “They have identifi ed the order in which diff erent mutations appear in melanoma and identified an intermediate type of lesion in this disease that is not as benign as a nevus, but not as dangerous as a melanoma”. She added that these results reaffirmed the value of sunscreen and removal of lesions at an early stage, before many mutations have accumulated. Bastian believes that the categories that emerged from the sequencing can be used to make risk assessment more objective and quantifiable. “It opens the door for precision medicine to improve diagnostic accuracy”, he concluded.

Atypical glandular cells and risk of cervical cancer

Abstract: www.thelancet.com/oncology Vol 17 March 2016 e96 Women with atypical glandular cells (AGC) identifi ed at cervical screening remain at heightened risk of developing invasive cervical cancer for up to 15·5 years compared with women with normal cytology, according to new research. The investigators examined the records of 3 024 340 Swedish women who had undergone cervical cytology testing from 1980 to 2011. 14 625 were diagnosed with AGC on their fi rst abnormal screening, of whom 65 633 were diagnosed with high grade squamous intraepithelial lesions (HSIL), and 244 168 were diagnosed with low grade squamous intraepithelial lesions (LSIL). Within 6 months of diagnosis, 1·4% of women with AGC, 2·5% of women with HSIL, and 0·2% of women with LSIL were found to have cervical cancer. For up to 15·5 years, the cumulative incidence of invasive cervical cancer, mainly adenocarcinoma, for women with AGC was 2·6% (95% CI 2·3–2·9). Beyond the fi rst 6 months, their risk of developing invasive cervical cancer remained higher than women with either HSIL or LSIL for up to 6·5 years. But, whereas 85·9% of women with HSIL, and 43·2% of women with LSIL, were followed up within 6 months with histological assessment, only 53·8% of women with AGC underwent such testing. “We need more attention and closer follow-up of women with AGC”, said co-author Par Sparen (Karolinska Institutet, Stockholm, Sweden). In view of the small number of aff ected women, such follow-up should be feasible. Sweden, with a population of about 10 million, sees around 1300 women diagnosed with AGC every year, of which 30–35 are likely to develop cervical cancer within 15 years. “We should be thinking of any glandular changes more in the way we think of high grade abnormalities than low grade ones”, agrees Peter Sasieni (Queen Mary University of London, London, UK). He cautioned that this might result in some over-treatment—glandular tumours can be tricky to fi nd. “There is certainly the question of what is the best way of sampling once you have seen something in the glandular cells”, added Sasieni. Another issue is that of diff erent risk profi les among women diagnosed with AGC. It is not clear whether the two glandular cytological categories in the British classifi cation system directly correspond to the Swedish AGC. Questions also remain over the clinical stage at which women with AGC tend to be diagnosed with cancer. “We have some limited staging data for the last 10 years, and we hope to look into this”, said Sparen.

Radiotherapy challenges in Uganda

Abstract: www.thelancet.com/oncology Vol 17 May 2016 e185 Uganda’s only external beam radiotherapy machine has recently broken down. The Cobalt-60 machine was donated to Mulago National Referral Hospital (Kampala, Uganda) by the International Atomic Energy Agency (IAEA) in 1995. A new machine has been ordered from the IAEA, but Jackson Orem (Uganda Cancer Institute, Kampala, Uganda), estimates it will be at least six months before it arrives. Around 2000 patients, many of whom travel from neighbouring countries, undergo radiotherapy at Mulago Hospital every year. “Having one country lose the ability to give radiotherapy has a multicountry, rippling effect throughout cancer care in sub-Saharan Africa”, points out Peter Kingham (Memorial Sloan Kettering, New York, NY, USA). “It really shows how tenuous the supply line and availability is”. Over half of the nations in Africa have no radiotherapy services at all. Where they do exist, services are scarce. Nigeria (population 180 million) usually has to manage with 2–4 functional radiotherapy machines. “Patients travel throughout the country to whichever machine happens to be working at the time”, said Kingham. Most of Uganda’s radiotherapy patients have cervical cancer. The country has one of the highest rates of the disease in the world, and 80% of patients present with late-stage disease. Radiotherapy is predominantly used as a palliative measure. “We have an elaborate system of palliative care that we can now fall back on”, said Orem. He added that for cervical cancer patients, clinicians can request treatment with Uganda Cancer Institute’s newly installed high dose bracytherapy equipment. A lack of radiotherapy means that women with breast cancer must have mastectomies instead of lumpectomies, but this is not contrary to common practice in Uganda. “In most developing countries, including Uganda, mastectomy is still the norm rather than the exception”, points out Orem. Uganda’s public healthcare system does provide for sending patients for treatment abroad if necessary, although the interrupted radiotherapy services are not likely to result in a sizeable exodus. Orem views the demise of the Cobalt-60 machine as a setback, rather than a catastrophe. The Uganda Cancer Institute has ambitious plans to establish bunkers containing advanced treatment technologies such as linear accelerators. “I am quite sure that in a short time the landscape for treatment for patients with cancer who require radiotherapy will be quite diff erent in Uganda”, Orem said.

Ipilimumab prolongs survival in advanced melanoma.

Abstract: Ipilimumab, a GLA-4 inhibitor, significantly prolongs overall survival in patients with advanced melanoma who have had com plete lymphnode dissection, according to a new study. The invest i gators assigned patients with stage III melanoma, who had under gone surgery within the previous 12 weeks, to receive either 10 mg/kg ipilimumab for 3 weeks for four doses, then every 3 months for up to 3 years (n=471) or the placebo (n=476). The primary endpoint was recurrence-free survival: at 5 years 40·8% of patients in the ipilimumab group were recurrence free compared with 30·3% of patients in the placebo group (HR 0·76 [95% CI 0·64–0·89; p<0·001). The proportion of patients who were free of distant metastases was also higher in the ipilimumab group (48·3%) compared with placebo (38·9%; HR 0·76 [95·8% CI 0·64–0·92]; p=0·002). Overall survival in the ipilimumab group was higher than in the placebo group: 65·4% versus 54·4% (hazard ratio [HR] 0·72 [95·1% CI 0·58–0·88]; p=0·001). “You see these benefi ts both in patients with sentinel-node positive disease, and with palpable nodal disease”, said fi rst author Alexander Eggermont (Gustave Roussy Cancer Campus Grand Paris, Villejuif, France). Five deaths were attributed to ipilimumab and over half the patients in the treatment group stopped taking it because of adverse events. 40% of patients did not go beyond four doses of ipilimumab, and 13% lasted the entire 3-year treatment schedule. Yet this did not aff ect survival, raising the question of whether four doses might be all that is needed. Paul Lorigan (Christie NHS Foundation Trust, Manchester, UK) suggested that ipilimumab’s toxicity is a surrogate for activity. “So the patients who are getting more toxicity are getting more benefi t, and they are getting this benefi t fairly early on”, he explained. “It is a question of managing the toxicity.” This is particularly important for adjuvant therapies, where the risk-benefi t calculations diff er from those involving therapies that target advanced, metastatic disease. Eggermont stresses the value of keeping in mind the fact that symptoms such as diarrhoea in patients taking ipilimumab might be drug-related. This could help facilitate prompt clinical intervention—most of the serious adverse events associated with ipilimumab can be eff ectively treated with corticosteroids. Two ongoing trials are examining the effi cacy of anti-PD-1 agents as adjuvant therapies for melanoma.

Postoperative radiotherapy for paediatric medulloblastoma.

Abstract: www.thelancet.com/oncology Vol 17 September 2016 e381 The proportion of children with medulloblastoma receiving deferred postoperative radiotherapy more than tripled in the USA from 2004 to 2012, according to new research. Benjamin Kann (Yale University School of Medicine, New Haven, CT, USA) and colleagues examined the records of 816 patients between 3 and 8 years of age drawn from the US National Cancer Database. The patients were diagnosed with medulloblastoma between 2004 and 2012 and had surgery and adjuvant chemotherapy. Those who were given radiotherapy at least 90 days after surgery were designated as having deferred postoperative radiotherapy. Overall, 123 (15%) of 816 patients had deferred radiotherapy. The main factor in determining whether treatment was given upfront was the patient’s age. 50 (36·8%) of 136 3-year-olds had deferred radiotherapy, compared with five (4·1%) of 123 8-year-olds. In 2005, 8% of patients had deferred radiotherapy. By 2012, this had risen to 27%. Yet the investigators found that deferring radiotherapy was strongly associated with poorer overall survival (hazard ratio 1·95 [95% CI 1·15–3·31]). Craniospinal radiotherapy is the type of radiotherapy necessary to treat medulloblastoma. If given to children younger than 3 years, it can have devastating eff ects, including severe learning diffi culties. Clinicians therefore avoid prescribing radiotherapy for this patient population. Deferring radiotherapy for slightly older children might reflect legitimate concerns driving a decision to avoid radiotherapy or delay it until the patient is better able to tolerate it. “But there needs to be an understanding that deferring radiotherapy may be coming at the cost of decreased survival”, explained Kann. “This paper underscores the fact if the decision is made to give a child radiotherapy, it should be given as soon as possible”, said Richard Gilbertson (University of Cambridge, Cambridge, UK). But he cautioned against over-interpreting the results. “Medulloblastoma is at least four diff erent types of disease; some patients with particular forms may only need a reduced dose or even no radiotherapy”, explained Gilbertson, adding, “This study views medulloblastoma as a single entity; it should not be used to support a blanket statement about treating all types of medulloblastoma.” Kann believes that the standard-ofcare for paediatric medulloblastoma should remain upfront postoperative radiotherapy, but agreed that deferral is likely to be appropriate for certain subgroups.

Progression-free survival with regorafenib in gastric cancer

Abstract: www.thelancet.com/oncology Published online June 30, 2016 http://dx.doi.org/10.1016/S1470-2045(16)30284-4 1 The oral multikinase inhibitor regorafenib improves progression-free survival for patients with refractory advanced gastric adenocarcinoma, according to new research by Nick Pavlakis (University of Sydney, Sydney, NSW, Australia) and colleagues. The multinational, randomised phase 2 trial assessed 147 patients with gastric adenocarcinoma who had previous undergone chemotherapy. 97 patients were randomly assigned to supportive care plus regorafenib, and 50 patients were assigned to supportive care plus placebo. Median progression-free survival (the primary endpoint) in the regorafenib group was 2·6 months (95% CI 1·8–3·1), compared with 0·9 months in the placebo group (0·9–0·9; hazard ratio 0·40 [95% CI 0·28–0·59], p<0·001). “Regorafenib was licensed on the basis of its use as a last-line drug for treating colon cancer”, explained coauthor David Goldstein (University of New South Wales, Sydney, NSW, Australia). “We found that it is an active drug for gastric cancer, and the survival benefit justifies additional research”. Results from the study, which included patients from Australia, Canada, New Zealand, and South Korea, showed that the progression-free survival benefi t was most profound in the Korean patients. “We do not know whether this is just by chance, or whether there is a pharmacokinetic or pharmacogenomic explanation. These are questions we hope to answer in [a] much larger phase 3 trial”, added Goldstein. Andrew Coveler (Seattle Cancer Care Alliance, Washington, DC, USA) welcomed the findings. “Any new therapy for gastric cancer, which is a hard to treat disease, would be a very good thing”, he said. But he cautioned that regorafenib is fairly toxic. 32% of patients in the regorafenib group in the study had at least one serious adverse event, including hypertension and increased aspartate aminotransferase, compared with 18% of patients in the placebo group. Finding a biomarker to identify patients who would benefi t from medication would minimise unnecessary treatment. “That has been the challenge in the angiogenesis fi eld for some years now”, pointed out Goldstein. Coveler believes that although VEGF inhibitors such as regorafenib will be important in treating gastric cancer, the future belongs to immunotherapy. “PD-1 or PD-L1 inhibitor therapy is coming to gastric cancer”, he says. “They have less toxicity, and retraining the immune system [could] give a prolonged and durable response”.

Oncologic Drugs Advisory Committee and conflicts of interest.

Abstract: www.thelancet.com/oncology Vol 17 March 2016 e95 Almost a third of public speakers at meetings of the US Oncologic Drugs Advisory Committee (ODAC) have a fi nancial relation with the company seeking drug approval, according to new research. The committee advises the Food and Drug Administration (FDA) on treatments for cancer. Its meetings typically include a presentation from the pharmaceutical company along with a public hearing. Matthew Abola and Vinay Prasad looked at the 103 public speakers called to discuss cancer drugs during the course of the ODAC’s 2009–14 meetings. They noted that 46 (45%) of 103 speakers had the cancer targeted by the drug being discussed. 32 (31%) speakers had used the drug. 58 (56%) speakers were associated with organisations that were relevant to either the drug or the cancer that it treated. 31 speakers (30%) had a financial conflict of interest—the company seeking drug approval might have covered travel expenses or funded the organisation represented by the speaker. 95 (92%) of the speakers argued in favour of approving the drug in question. “The US FDA does not always convene an ODAC”, explains Prasad (Oregon Health and Science University, Portland, OR, USA). “They do so often in cases where the risk-benefi t is even more questionable than usual, and even in this setting, the drug industry appears to try to control the outcome”. He argues that while the industry sponsors patient advocacy groups and individual speakers, patients who are concerned about overall survival, side-effects, and costs go unrepresented. “It seems that the FDA is only seeing one very small sliver of public opinion, and that sliver is potentially infl uenced by fi nancial relationships with organisations interested in the outcome”, notes Aaron Kesselheim (Harvard University, Cambridge, MA, USA). “If the FDA is going to seek direct public opinion in these sort of cases, it must reach out and make sure it is really hearing the true spectrum of opinions”. One possibility would be to limit the number of slots available to speakers with industry ties. Prasad has a more radical idea. “The ideal solution would be for patients on pivotal trials to record video diaries”, he said. ODAC meetings would include a selection of randomly chosen videos. “Sometimes, it is important to hear from patients who died and cannot make the meeting—it doesn’t make sense just to hear from people who did well”, he concluded.

Molecular subgroups of pancreatic cancer

Abstract: www.thelancet.com/oncology Vol 17 April 2016 e139 Sales in 2016 in the USA of the top 20 cancer drugs dosed by bodyweight and produced in single dose vials are projected to be US$18 billion. But manufacturers are doing little to ensure that vial size minimises leftover product, and a new analysis estimates that these drug leftovers will cost the USA $1·8 billion this year. Leonard Saltz (Memorial Sloan Kettering Cancer Centre, New York, NY, USA), a co-author of the study, cites the example of pembrolizumab. The drug is sold in Europe in 50 mg vials, but the manufacturers withdrew this size from the USA market and replaced it with a 100 mg vial. “That is very large packaging for a drug that costs in excess of $50 per mg”, Saltz said. For example, patients typically require a 140 mg dose of pembrolizumab, so two 100 mg vials would be needed to make up this dose, which means that 60 mg of the drug would be left over. Saltz and colleagues estimate that, over the course of 2016, the cost of wasted pembrolizumab will add up to $198 million. The proportion of leftover product ranged from 1% to 33% for the 20 examined drugs. Bendamustine is available in vials of four diff erent sizes, meaning that there are only two unavailable multiples of 5 in the typical adult dosage range of 110–330 mg. Bendamustine’s estimated contribution to total waste is a relatively small $7·4 million. If other manufacturers provided drugs in vial sizes that similarly minimised waste, the investigators believe that the cost of leftover drugs could be reduced to $400 million. “In the USA, we have lost all connection between the money that we are spending and the value that we expect to receive from it”, adds Saltz. He advocates regulation to limit permissible levels of wasted drugs, or a system in which only those drugs that are used are paid for. “There is careful documentation of how much drug is given to each patient; it would be quite easy to do the billing and reimbursement on that basis”, Saltz concluded. Vinay Prasad (Oregon Health and Science University, Portland, OR, USA) believes that the problem has deeper roots. “The cost of making the drug and the cost of the drug have little connection”, Prasad explained. So smaller vials need not necessarily lead to smaller costs—Prasad expects that manufacturers would respond by raising drug prices. In which case, savings would be trickier to realise, but such a system might have greater transparency.

NICE restructures the Cancer Drugs Fund.

Abstract: National Health Service (NHS) England and the National Institute for Health and Care Excellence (NICE) have unveiled proposals for restructuring the Cancer Drugs Fund. After its scheduled expiration on March 31, 2016, the fund will become part of the NICE appraisal process. It will be a transitional fund, designed to offer patients with cancer access to promising drugs for which insuffi cient evidence exists for a recommendation for routine commissioning. In the new system, NICE will issue draft guidance for cancer drugs before market authorisation and again within 90 days of authorisation. They will recommend some drugs for routine use, reject use of others, and recommend some for use in the new fund. Unlike the current version, the proposed fund will collect data on patient outcomes, usually for up to 24 months, after which NICE will issue a recommendation on whether the drug should become part of routine commissioning or be available only on request. At this point, irrespective of NICE’s recommendation, a drug will exit the fund. Cancer Research UK’s Emlyn Samuel cautiously welcomed the proposals. “The crucial thing is that it looks like the decision-making on what drugs should enter the Cancer Drugs Fund evaluation, and on whether the drugs become routinely available on the NHS, is made by NICE”, he told The Lancet Oncology. The present fund, fi rst established in 2010, has run as a parallel system, allowing some pharmaceutical companies to bypass NICE. Samuel continued, “it is a good move to return decision-making responsibility to NICE, because they have the expertise to do this well”. The Cancer Drug Fund, which has provided treatment to more than 72 000 patients, was set up with a budget of £200 million. This budget has since been increased twice. For the year 2014–15, it overspent by 48%. Its budget for 2015–16 is £340 million. NHS England and NICE did not specify the likely budget for the restructured fund. But converting the fund into a fi xed cost might help the NHS fi nd a small amount of the £22 billion of effi ciency savings the Government expects it to deliver by 2020–21. The proposals will now undergo a 12-week public consultation, to end on Feb 11, 2016. The new Cancer Drugs Fund is expected to become operational from April 1, 2016.

Incidence of breast cancer in women with type 2 diabetes

Abstract: www.thelancet.com/oncology Published online October 27, 2016 http://dx.doi.org/10.1016/S1470-2045(16)30555-1 1 Findings from a population-based cohort study have shown that glucagon-like peptide-1 (GLP-1) analogues for treatment of type 2 diabetes are not associated with an increased incidence of breast cancer. Blánaid M Hicks (Jewish General Hospital, Montreal, QC, Canada) and colleagues examined the records of 44 984 women who were newly treated with glucose-lowering drugs from Jan 1, 2007, to March 31, 2015. The women were drawn from Clinical Practice Research Datalink, a UK database, and were all aged 40 years or older. Mean follow-up was 3·5 years (SD 2·2). 2473 (5·5%) women were prescribed GLP-1 analogues, and 8848 (19·7%) were prescribed dipeptidylpeptidase-4 (DPP-4) inhibitors during the study period. Compared with DPP-4 inhibitors, GLP-1 analogues did not increase the incidence of breast cancer (hazard ratio 1·40 [95% CI 0·91–2·16]). Women using GLP-1 analogues for 2–3 years have an increased incidence of breast cancer (hazard ratio 2·66 [95% CI 1·32–5·38]) compared with those who took the drug for a diff erent amount of time (1·21 [0·64–2·30] for ≤1 year of use, 1·18 [0·59–2·38] for 1·1–2 years of use, and 0·98 [0·24–4·03] for >3 years of use). “A likely explanation is detection bias”, explained senior author Laurent Azoulay (McGill University, Montreal, QC, Canada). But he added that a tumour promoter eff ect could not be ruled out. Lorraine Lipscombe (University of Toronto, ON, Canada) noted that the study did not show a dose–response eff ect. “Normally if a drug increases the risk of cancer, then the longer patients are on the drug, the higher the risk”, she explained. “Here, the risk peaked early and then dropped off —that tells me there must be something else going on.” Anna Morris from Diabetes UK (London, UK) agreed. “There is absolutely nothing in this paper that would suggest that people should stop taking these medications”, she told The Lancet Oncology. Obesity is a risk factor for both type 2 diabetes and breast cancer. It has been suggested that the weight loss associated with GLP-1 analogue use can lead to improved detection of breast lumps. “It was not possible to evaluate the eff ect of rapid weight loss on the detection of breast cancer in this study”, said Azoulay. “It would be of interest to look at this using data from recent large randomised controlled trials of GLP-1 analogues, where bodyweight was measured before randomisation and during follow-up.”