放眼全球看待癌症

Think globally about cancer

Nat Med. 2019 Mar. IF=87.241

Think globally about cancer - PubMed (nih.gov)

Cancer surveillance programs have reported a global downward trend in cancer mortality rates for most common tumor types. However, startling geographic inequalities exist, and some cancers continue to pose a challenge. Ensuring global access to high-quality diagnostic and treatment approaches is needed to make decreasing cancer deaths a more widespread trend

In January, the AmericanCancer Society releasedthe results of its yearly cancersurveillance analysis, which revealed a continuing decline in cancer mortality rates in the United States (CA Cancer J. Clin. 69, 7–34, 2019). Results from the report were in line with trends describedin the last CONCORD3worldwide surveillanceprogram, which included data on millionsof patients across more than 70 countries between 2000 and 2014 and encompassed 18 different cancer typesthat represent 75% of the globalcancer burden (Lancet391, P1023–1075, 2018). As compared with previouspopulation-based surveillancestudies by the CONCORD programme, coordinated by the LondonSchool of Hygiene&TropicalMedicine,globalmetrics for this last surveyedperiod show that the 5-year survival is highest in most common adult cancers, including colorectal cancer (50–70%) and breast cancer (70–85%). These survivalrates are considered to be the fruitsof increased awareness,improved screening and prevention programs and better treatments (such as the dramaticincrease in survival in HER2-positivebreast cancer patients sincetrastuzumab was introduced in1998).

However, the CONCORD3study also highlightsthe finding that improvementsin survival rates for most of these commonadult cancers are markedly differentin high-income countries as compared with lessaffluent regions. The picture is similar for pediatric cancers. Survivalrates for children diagnosed with brain canceror acute lymphoblastic leukemia(ALL) are within 80% in countriessuch as Denmark,whereas these rates remain below 40% for pediatricpatients with brain cancerin Brazil. Such huge differences can be attributedin part to inequalities in accessto cutting-edge therapies— which can often be cost-prohibitive for the overall population or be unaffordable to some national healthcare systems — and to a shortage of population screening programsin areas with more limitedresources. Recently announced initiatives are expected to help breach such disparities. These include the effort by the UnitedStates Agency for International Development (USAID) and US.National Academies of Science to implement cervical cancer prevention programs in sub-Saharan Africa,and the campaign launched by UnitedKingdom’s National Health System to bring mobile lung screening units to most deprived regionsin that country.

An equally relevant issue thatneeds to be solved before strategies to bridge these persistent survivalinequalities can be put in place is the lack of surveillance data for a large number of lower-middle- and low-income countries. Recent advances in leveraging the wealth of data collected from mobile technologies and wearablesensors have the potential to deliver reliable health readouts in real-time that can be harnessed tofill some of these information gaps.

Similarly, the fast-paced implementation of artificial intelligence approaches in healthcare is enabling improved screening and early cancer detectionprocedures ataffordable costs. Some examplesinclude the application of computer vision for identifying skin cancer from molephotographs, which can be taken using a smart-phonecamera (Nature 542, 115–118, 2017), and enhancement of the diagnostic accuracy of malignant of pulmonarynodules imaged in routine diagnostic chestscans (Sci. Rep. 7, 46479, 2017). Applications of machine learning in the fieldof liquid biopsiesallowing for early detection of pancreatic, ovarianand other cancers

based on cell free circulating DNA analyses (Nature 563, 579–583, 2018) also holds promise in democratizing early diagnosisat lower costs. Although thesestudies havebeen so far performedretrospectively, they describesensitive and cost-efficient approaches that, if confirmedin prospective trials and adaptedto cost-efficient interfaces,have the potential to optimizethe clinical management of patients with cancer and improve outcomes at a global scale.

Institutional policies and research efforts by the biomedical community are helping to reduce mortality from most cancers. However, there are still some tumor types for which tilting the scales remains difficult, regardless of geographical location. For example, survival rates have stayed flat since the late 1990sfor ovarian cancer(30–50%) and adult brain cancer (20–40%; and under 10% in the case of glioblastoma), and also remain low for pancreatic cancer (5–15%). These cancer types present atough challenge for doctors and biomedical researchers alike — due in part to biological hurdles that prevent the delivery of effective therapies.

Ovarian or pancreatic cancersare rarely caught early because of a lack of screening strategies and vague symptoms that only become prominentat advanced stages when tumors have spread or are refractory to treatments, and so the need for early detection approaches in these tumor types is most urgent.

With cancer diagnosesincreasing globally each year,it is essential that campaigns for prevention and early cancer detectionare implemented, especially in resource-limited countries where cancer hasmore drastic effects on patients. Granting equal access to affordable high-quality care should bridgethe gaps in global cancer survival. With the advent of innovative cost-effective technologies and improved treatment approaches, the hope is thatdeclining cancer mortality rates will reach all corners of the globe.