Rak płuca
Rokowania, prognozy i postęp choroby
Rak płuca pozostaje jednym z najczęściej diagnozowanych nowotworów z 5-letnim wskaźnikiem przeżycia na poziomie 20,5%. Wczesne wykrycie, zwłaszcza za pomocą niskodawkowej tomografii komputerowej (LDCT), może zmniejszyć śmiertelność o 20-24% u osób wysokiego ryzyka. Modele predykcyjne, takie jak PLCOM2012 (AUC około 0,80), umożliwiają skuteczną identyfikację pacjentów do badań przesiewowych, co przekłada się na wykrywanie 76% nowotworów w stadium I-II. Indywidualizacja odstępów między badaniami oraz wykorzystanie biomarkerów, w tym płynnej biopsji (CTC, ctDNA), poprawiają diagnostykę, monitorowanie progresji i odpowiedź na terapię. Algorytmy oparte na rutynowych badaniach krwi (indeks RBT) oraz techniki głębokiego uczenia wspierają prognozowanie i wczesną diagnostykę, szczególnie w warunkach ograniczonych zasobów.
- Prognozy raka płuca w badaniach przesiewowych
- Biomarkery i płynna biopsja w prognozowaniu raka płuca
- Sztuczna inteligencja i głębokie uczenie w prognozowaniu raka płuca
- Wyzwania i ograniczenia prognozowania w badaniach przesiewowych raka płuca
- Niedoszacowanie ryzyka w modelach predykcyjnych
- Luki w modelach dla osób niepalących
- Ograniczenia istniejących modeli prognostycznych
- Zalecenia i wytyczne dotyczące badań przesiewowych
Prognozy raka płuca w badaniach przesiewowych
Rak płuca jest jednym z najczęściej diagnozowanych nowotworów i główną przyczyną zgonów związanych z rakiem na całym świecie 1. Choroba ta charakteryzuje się ogólnie niekorzystnym rokowaniem, z 5-letnim wskaźnikiem przeżycia wynoszącym jedynie 20,5% 2. Jednakże wczesne wykrycie raka płuca znacząco poprawia rokowanie pacjenta i zwiększa szanse na skuteczne leczenie 3. Badania przesiewowe w kierunku raka płuca za pomocą niskodawkowej tomografii komputerowej (LDCT) mogą zmniejszyć śmiertelność z powodu tego nowotworu o 20% 45, a w niektórych badaniach nawet o 24% u mężczyzn wysokiego ryzyka 6.
Modele predykcji ryzyka w badaniach przesiewowych
Liczne modele predykcyjne zostały opracowane w celu identyfikacji osób o wysokim ryzyku zachorowania na raka płuca, aby poprawić wczesne wykrywanie i wskaźniki przeżycia 7. W ostatnich latach zaproponowano podejście oparte na indywidualnym ryzyku do zarządzania wynikami badań przesiewowych raka płuca. Takie podejście uwzględnia indywidualne czynniki ryzyka oraz cechy obrazu LDCT w obliczeniach natychmiastowego i przyszłego (rocznego) ryzyka wykrycia raka płuca 8.
Model PLCOM2012 jest jednym z najszerzej walidowanych i najskuteczniejszych modeli predykcyjnych. W metaanalizie wykazano, że spośród ośmiu modeli, które przeszły obszerne zewnętrzne walidacje, PLCOM2012 wykazał najlepszą ogólną dyskryminację predykcyjną w zewnętrznych kohortach 9. Model ten osiągnął wartość AUC 0,803 w zestawie danych rozwojowych (ramię kontrolne PLCO) i 0,797 w zestawie walidacyjnym (ramię interwencyjne PLCO) 10.
W badaniach przeprowadzonych w Manchester Lung Health Check, model PLCOm2012 skutecznie klasyfikował większość przypadków raka płuca w grupie wysokiego ryzyka poddanej badaniom przesiewowym. W tej kohorcie 91% (n=105/115) przypadków raka płuca wystąpiło w grupie wysokiego ryzyka poddanej badaniom przesiewowym, a trzy czwarte z tych nowotworów zostało zdiagnozowanych we wczesnym stadium (76% w stadium I-II) 11. Jest to znacząca poprawa w porównaniu do krajowej dystrybucji stadiów przed wdrożeniem badań przesiewowych, gdzie mniejszość przypadków jest diagnozowana we wczesnym stadium 12.
Efektywność modeli opartych na ryzyku
Wykorzystanie zindywidualizowanych modeli ryzyka do określenia, kto powinien być skierowany na obrazowanie w krótkim odstępie czasu i do wyboru długości odstępów między badaniami przesiewowymi, może znacznie zwiększyć efektywność programów badań przesiewowych raka płuca 13. Porównanie z protokołem National Lung Screening Trial (NLST) wykazało, że protokół NHS England skierowałby o 36% mniej osób z nieprawidłowymi wynikami badań na obrazowanie w krótkim odstępie czasu (np. 3-miesięcznym) lub przyspieszone obrazowanie nadzorcze, przy jednoczesnym opóźnieniu diagnozy jedynie dla 4% nowotworów 14.
Przy zastosowaniu modelu LCRAT+CTpos dla natychmiastowego ryzyka w celu identyfikacji osób do obrazowania w krótkim odstępie czasu, próg ryzyka 0,60% również opóźniłby diagnozę dla 4% nowotworów, ale skierowałby o 42% mniej osób z nieprawidłowymi wynikami badań na obrazowanie w krótkim odstępie czasu w porównaniu z NLST. Jest to dodatkowa 9% redukcja w porównaniu z protokołem NHS England 15.
Programy zalecające dwuletnie badania przesiewowe dla większości uczestników mogłyby znacznie poprawić czułość, wykorzystując modele ryzyka do identyfikacji osób wysokiego ryzyka, które powinny być kierowane na coroczne badania przesiewowe 16.
Biomarkery i płynna biopsja w prognozowaniu raka płuca
Wzrasta znaczenie biomarkerów w prognozowaniu przebiegu raka płuca. Panel biomarkerów oparty na badaniu krwi znacznie poprawia ocenę ryzyka raka płuca dla badań przesiewowych 17. Płynna biopsja pojawia się jako obiecująca metoda identyfikacji pacjentów z wysokim ryzykiem progresji choroby po leczeniu chirurgicznym, a także do longitudinalnego monitorowania progresji choroby i odpowiedzi na terapię 18.
Rutynowe badania krwi i biomarkery zapalne
Istnieją dowody, że stan zapalny, immunologiczny i metaboliczny jest związany z przeżyciem pacjentów z rakiem. Badacze opracowali prosty algorytm do przewidywania wyników raka płuca na podstawie okołooperacyjnych rutynowych badań krwi (RBT). Profile zapalne, immunologiczne i metaboliczne zostały wykorzystane do stworzenia pojedynczego algorytmu (indeks RBT) przewidującego przeżycie pacjentów z rakiem płuca 19.
Pacjenci z wysokim indeksem RBT mieli wyższe 5-letnie ryzyko zgonu niż pacjenci z niskim RBT (skorygowany HR 1,93, 95% CI 1,62-2,31), podobnie jak pacjenci z pośrednim poziomem (1,34, 1,11-1,61) 20. Indeks RBT był w stanie przewidzieć 30- i 90-dniową śmiertelność, a także 5-letnie przeżycie w różnych warunkach onkologii klatki piersiowej 21.
Płynna biopsja w diagnostyce i monitorowaniu
Płynna biopsja otwiera nowe możliwości we wczesnym badaniu przesiewowym, diagnostyce i leczeniu raka płuca, szczególnie gdy nie można uzyskać próbek tkanki 22. Markery nowotworowe przyczyniają się do wczesnej diagnozy raka płuca i odgrywają ważną rolę we wczesnym wykrywaniu i leczeniu, a także w medycynie precyzyjnej, zindywidualizowanym leczeniu oraz przewidywaniu rokowania 23.
W płynnej biopsji analizowane są różne komponenty:
- Krążące komórki nowotworowe (CTC) – wykrywanie CTC zostało zgłoszone jako alternatywna metoda u pacjentów z rearanżacją genu kinazy chłoniaka anaplastycznego (ALK) w NSCLC, gdy nie można uzyskać odpowiednich próbek biopsji tkankowej 24. DNA wykrywane przez CTC może przewidywać progresję raka płuca i wtórną oporność na leki 25.
- Krążący DNA guza (ctDNA) – wykrywanie ctDNA jest używane do kierowania wyborem odpowiednich inhibitorów kinazy tyrozynowej ALK (ALK-TKI), co może dostarczyć cennych informacji do precyzyjnego leczenia pacjentów z rakiem płuca, szczególnie dla pacjentów, którzy nie kwalifikują się do biopsji 26. Analiza ctDNA wykazała wysoką wartość we wczesnej diagnostyce i prognozie 27.
Ciągłe monitorowanie odpowiedzi terapeutycznej raka płuca za pomocą analizy CTC ma dużą wartość i perspektywę badawczą 28.
Sztuczna inteligencja i głębokie uczenie w prognozowaniu raka płuca
Algorytmy głębokiego uczenia mają potencjał do zmiany klinicznego przepływu pracy w raku płuca 29. Zaawansowane techniki obliczeniowe są coraz częściej wykorzystywane do poprawy dokładności diagnostycznej i prognostycznej w badaniach przesiewowych raka płuca.
Systemy wspomagania decyzji klinicznych
W prospektywnej kohorcie badań przesiewowych raka płuca przeprowadzonej w dwóch obszarach wiejskich zachodnich Chin, system głębokiego uczenia został opracowany do automatycznego wykrywania guzków (recall 0,9507; FROC 0,6470) i stratyfikacji ryzyka (ACC 0,8696; makro-AUC 0,8516) 30. To nowatorskie podejście w zastosowaniach medycznych może pomóc klinicystom skutecznie ułatwić wczesną diagnozę raka płuca, szczególnie w miejscach o ograniczonych zasobach 31.
Pacjenci z wczesnym stadium raka płuca, którzy otrzymali leczenie z intencją wyleczenia, mają znacznie lepsze rokowanie w porównaniu z tymi, u których rak płuca jest w zaawansowanym stadium 32.
Modele predykcyjne oparte na danych rejestrowych
W badaniu populacyjnym przeprowadzonym w Danii opracowano i zwalidowano modele predykcyjne do szacowania indywidualnego ryzyka zachorowania na raka płuca w ciągu 1 roku dla wszystkich osób w wieku 40 lat lub starszych mieszkających w Danii na dzień 1 stycznia 2017 r. 33. Najlepszy model dla osób bez wcześniejszego rozpoznania raka w ciągu dziesięciu lat uwzględniał zarówno kody ICD-10, kody ATC, liczbę kontaktów z lekarzem pierwszego kontaktu i specjalistami, procedury oraz podstawowe cechy socjodemograficzne 34.
Walidacja modelu wykazała rozsądną wartość predykcyjną z AUC 0,80 w populacji bez wcześniejszego rozpoznania raka w porównaniu z innymi modelami predykcyjnymi raka płuca 35. Takie modele predykcyjne oparte na rejestrach mogą wspierać klinicystów i planistów opieki zdrowotnej w identyfikacji osób zagrożonych rakiem płuca, które powinny być skierowane na dalsze badania 36.
Wyzwania i ograniczenia prognozowania w badaniach przesiewowych raka płuca
Niedoszacowanie ryzyka w modelach predykcyjnych
W badaniu Manchester Lung Health Check zaobserwowano, że modele ryzyka generalnie niedoszacowywały ryzyko, niezależnie od statusu badań przesiewowych 37. Ogólnie o 44% więcej przypadków raka płuca zostało zdiagnozowanych niż oczekiwano przy użyciu PLCOm2012, który przewiduje 6-letnie ryzyko raka płuca bez interwencji przesiewowej 38.
Niedoszacowanie ryzyka zaobserwowano zarówno w grupie wysokiego ryzyka poddanej badaniom przesiewowym, jak i w grupie niskiego ryzyka niepoddanej badaniom, chociaż przypadki raka w grupie niepoddanej badaniom nie były zgrupowane tuż poniżej progu kwalifikacyjnego 39. Większość modeli predykcyjnych analizowanych w tym badaniu niedoszacowywała ryzyko (z wyjątkiem LLPv2), a dyskryminacja w grupie poddanej badaniom przesiewowym wahała się w zakresie AUC 0,61-0,79 40.
Luki w modelach dla osób niepalących
Większość zachodnich modeli została opracowana wśród osób palących obecnie lub w przeszłości, ze względu na dobrze ustaloną zależność między paleniem a rakiem płuca 41. Istnieje jednak znacząca luka w modelach predykcyjnych dla osób niepalących 42.
Tylko niewielka część badań stratyfikowała uczestników na podstawie statusu palenia, a tylko dwa azjatyckie badania skupiały się na opracowaniu modelu wyłącznie wśród osób niepalących 43. Mimo to wczesne wykrycie raka płuca u osób niepalących jest ważnym priorytetem zdrowia publicznego 44.
Obecnie osoby niepalące nie kwalifikują się do badań przesiewowych w kierunku raka płuca poza niektórymi krajami Azji Wschodniej 45. Wraz ze wzrostem proporcji osób niepalących wśród przypadków raka płuca, istnieje pilna potrzeba opracowania metod identyfikacji osób wysokiego ryzyka wśród osób niepalących i włączenia ich do programów badań przesiewowych 46.
Ograniczenia istniejących modeli prognostycznych
Wydajność modeli związanych z genami nie uległa wyraźnej poprawie 47. Większość powszechnych modeli jest wysoce stronnicza, a odwołanie się do narzędzia PROBAST na początku badania może znacznie kontrolować stronniczość 48.
Istniejące modele powinny być walidowane w dużym zewnętrznym zestawie danych, aby dokonać znaczącego porównania 49. W porównaniu z systemem klasyfikacji TNM, model predykcji rokowania może poprawić dokładność i kierować spersonalizowaną terapią poprzez kombinację wielu czynników prognostycznych 50.
System klasyfikacji TNM, skala WHO-PS i klasyfikacja patologiczna powinny być włączone do wszystkich modeli 51. Obecny model predykcji rokowania niedrobnokomórkowego raka płuca (NSCLC) jest obciążony wysokim ryzykiem stronniczości, a promocja stosowania PROBAST może poprawić tę sytuację 52.
Zalecenia i wytyczne dotyczące badań przesiewowych
Amerykańska Grupa Zadaniowa ds. Usług Prewencyjnych (USPSTF) zaleca coroczne badania przesiewowe w kierunku raka płuca za pomocą LDCT u dorosłych w wieku od 50 do 80 lat, którzy mają 20-paczkolat historii palenia i obecnie palą lub rzucili palenie w ciągu ostatnich 15 lat 53. USPSTF zaleca przerwanie badań przesiewowych, gdy osoba nie paliła przez 15 lat lub rozwija problem zdrowotny, który znacznie ogranicza oczekiwaną długość życia lub zdolność lub chęć poddania się operacji raka płuca 54.
Umiarkowana korzyść netto z badań przesiewowych zależy od ograniczenia badań przesiewowych do osób o wysokim ryzyku, dokładności interpretacji obrazu podobnej lub lepszej niż w badaniach klinicznych oraz rozwiązania większości wyników fałszywie dodatnich za pomocą seryjnego obrazowania, a nie procedur inwazyjnych 55.
Rekomendacje dotyczące częstotliwości badań
Centra Medicare i Medicaid Services (CMS) określiły, że dowody są wystarczające, aby dodać wizytę doradczą dotyczącą badań przesiewowych w kierunku raka płuca i wspólnego podejmowania decyzji, oraz, dla odpowiednich beneficjentów, coroczne badania przesiewowe w kierunku raka płuca za pomocą LDCT, jako dodatkową korzyść z usługi prewencyjnej w ramach programu Medicare 56.
USPSTF zaleca coroczne badania przesiewowe w kierunku raka płuca za pomocą niskodawkowej tomografii komputerowej u dorosłych w wieku od 55 do 80 lat, którzy mają 30-paczkolat historii palenia i obecnie palą lub rzucili palenie w ciągu ostatnich 15 lat 57. Jest to zgodne z wynikami badania NLST, które wykazało, że osoby w wieku od 55 do 74 lat z historią intensywnego palenia mają o 20% mniejsze prawdopodobieństwo śmierci z powodu raka płuca, jeśli są badane za pomocą niskodawkowej spiralnej tomografii komputerowej niż standardowymi badaniami rentgenowskimi klatki piersiowej 58.
Interpretacja wyników badań przesiewowych
Jeśli na badaniu przesiewowym w kierunku raka płuca nie zostaną wykryte żadne nieprawidłowości, lekarz może zalecić wykonanie kolejnego skanu za rok. Pacjent może rozważyć kontynuowanie corocznych skanów, dopóki on i jego lekarz nie uznają, że nie przyniosą one prawdopodobnie korzyści, na przykład jeśli rozwinie inne poważne problemy zdrowotne 59.
Badania przesiewowe w kierunku raka płuca niosą ze sobą kilka ryzyk, takich jak znalezienie raka, który jest zbyt zaawansowany, aby go wyleczyć. Zaawansowane raki płuc, takie jak te, które się rozprzestrzeniły, mogą nie reagować dobrze na leczenie, więc znalezienie tych raków w badaniu przesiewowym może nie poprawić ani nie wydłużyć życia pacjenta 60.
Oparte na solidnych dowodach, badania przesiewowe za pomocą rentgena klatki piersiowej i/lub cytologii plwociny nie zmniejszają śmiertelności z powodu raka płuca w populacji ogólnej ani u osób palących kiedykolwiek 61. Biorąc pod uwagę obfitość i spójność dowodów, a także brak korzyści zaobserwowanych w badaniu PLCO, właściwe jest stwierdzenie, że badania przesiewowe w kierunku raka płuca za pomocą rentgena klatki piersiowej i/lub cytologii plwociny, niezależnie od płci lub statusu palenia, nie zmniejszają śmiertelności z powodu raka płuca 62.
Kolejne rozdziały
Zapraszamy do dalszego czytania naszego leksykonu.
Wybierz kolejny rozdział z menu poniżej, aby otworzyć nową podstronę kompedium wiedzy i uzyskać szczegółowe informację o leku, substancji lub chorobie.
Materiały źródłowe
- #1 Risk prediction models for lung cancer in people who have never smoked: a protocol of a systematic review | Diagnostic and Prognostic Research | Full Texthttps://diagnprognres.biomedcentral.com/articles/10.1186/s41512-024-00166-4
Lung cancer is one of the most commonly diagnosed cancers and the leading cause of cancer-related death worldwide. […] Despite this alarming trend, this population is ineligible for lung screening. […] With the increasing proportion of people who have never smoked among lung cancer cases, there is a pressing need to develop prediction models to identify high-risk people who have never smoked and include them in lung cancer screening programs. […] The results of the review will identify risk prediction models for lung cancer in people who have never smoked. These will be useful for researchers planning to develop novel prediction models, and for clinical practitioners and policy makers seeking guidance for clinical decision-making and the formulation of future lung cancer screening strategies for people who have never smoked.
- #2 Recommendation: Lung Cancer: Screening | United States Preventive Services Taskforcehttps://www.uspreventiveservicestaskforce.org/uspstf/recommendation/lung-cancer-screening
Lung cancer has a generally poor prognosis, with an overall 5-year survival rate of 20.5%. However, early-stage lung cancer has a better prognosis and is more amenable to treatment. […] The US Preventive Services Task Force (USPSTF) concludes with moderate certainty that annual screening for lung cancer with LDCT has a moderate net benefit in persons at high risk of lung cancer based on age, total cumulative exposure to tobacco smoke, and years since quitting smoking. […] The USPSTF recommends annual screening for lung cancer with LDCT in adults aged 50 to 80 years who have a 20 pack-year smoking history and currently smoke or have quit within the past 15 years. […] The USPSTF recommends discontinuing screening once a person has not smoked for 15 years or develops a health problem that substantially limits life expectancy or the ability or willingness to have curative lung surgery. […] The moderate net benefit of screening depends on limiting screening to persons at high risk, the accuracy of image interpretation being similar to or better than that found in clinical trials, and the resolution of most false-positive results with serial imaging rather than invasive procedures.
- #3 Recommendation: Lung Cancer: Screening | United States Preventive Services Taskforcehttps://www.uspreventiveservicestaskforce.org/uspstf/recommendation/lung-cancer-screening
Lung cancer has a generally poor prognosis, with an overall 5-year survival rate of 20.5%. However, early-stage lung cancer has a better prognosis and is more amenable to treatment. […] The US Preventive Services Task Force (USPSTF) concludes with moderate certainty that annual screening for lung cancer with LDCT has a moderate net benefit in persons at high risk of lung cancer based on age, total cumulative exposure to tobacco smoke, and years since quitting smoking. […] The USPSTF recommends annual screening for lung cancer with LDCT in adults aged 50 to 80 years who have a 20 pack-year smoking history and currently smoke or have quit within the past 15 years. […] The USPSTF recommends discontinuing screening once a person has not smoked for 15 years or develops a health problem that substantially limits life expectancy or the ability or willingness to have curative lung surgery. […] The moderate net benefit of screening depends on limiting screening to persons at high risk, the accuracy of image interpretation being similar to or better than that found in clinical trials, and the resolution of most false-positive results with serial imaging rather than invasive procedures.
- #4 NCA – Screening for Lung Cancer with Low Dose Computed Tomography (LDCT) (CAG-00439N) – Decision Memohttps://www.cms.gov/medicare-coverage-database/view/ncacal-decision-memo.aspx?proposed=N&NCAId=274
The NLST showed that people aged 55 to 74 years with a history of heavy smoking are 20 percent less likely to die from lung cancer if they are screened with low dose helical CT than with standard screening chest x-rays. […] The NLST demonstrated benefit by enrolling a large number of high exposure patients (smoking history) to be followed for several years to detect a significant decrease (247 deaths from lung cancer per 100,000 person-years in the low-dose CT group and 309 deaths per 100,000 person-years in the radiography group; number needed to screen (NNS) to prevent one lung cancer death = 320). […] Overall, based primarily on the results of the NLST, we find that the evidence is sufficient to determine that screening for lung cancer with LDCT is reasonable and necessary for the prevention or early detection of illness or disability for the exact population studied in the NLST. […] Given the burden of lung cancer on the United States population, a suitable screening test for lung cancer has been sought for many years. Lung cancer screening has been recommended by the USPSTF with a grade B recommendation for certain individuals.
- #5 Could lung cancer prognosis be predicted from a drop of blood? – Oncology Centralhttps://www.oncology-central.com/can-lung-cancer-prognosis-be-predicted-from-a-drop-of-blood/
A blood-based biomarker panel significantly improves lung cancer risk assessment for lung cancer screening. […] a blood-based protein panel plus prediction model shows potential to lower lung cancer death by identifying individuals who need CT screening. […] The only approved screening test for lung cancer is low dose computed tomography (LDCT) and it has been previously reported that LDCT reduces lung cancer mortality by 20%.
- #6 Lung Cancer Screening (PDQ®) – NCIhttps://www.cancer.gov/types/lung/hp/lung-screening-pdq
Two randomized trials have reported statistically significant reductions in lung cancer mortality associated with low-dose computed tomography (LDCT) screening. One trial reported that screening higher-risk individuals (30+ pack-years and either current smokers or quit within the past 15 years) aged 55 to 74 years three times, once annually, with LDCT reduced lung cancer mortality by 20% (95% confidence interval [CI], 6.8%26.7%; P = .004) and all-cause mortality by 6.7% (95% CI, 1.2%13.6%; P = .02) over screening with chest radiographs. An updated analysis showed that the estimated reduction in lung cancer mortality was 16% (95% CI, 5%25%). The other trial reported that among high-risk current and former smokers, men who were randomly assigned to four rounds of LDCT screening had a 24% reduction (95% CI, 6%39%) in lung cancer mortality, compared with men who were randomly assigned to no screening.
- #7 Predictive performance of risk prediction models for lung cancer incidence in Western and Asian countries: a systematic review and meta-analysis | Scientific Reportshttps://www.nature.com/articles/s41598-024-83875-6
Numerous prediction models have been developed to identify high-risk individuals for lung cancer screening, with the aim of improving early detection and survival rates. […] Extensive randomized trials, including the 2013 National Lung Screening Trial (NLST) and the 2020 NELSON trial, have demonstrated the efficacy of low-dose computed tomographic (LDCT) screening as an effective tool for early lung cancer detection. […] Our review also highlights a significant gap in prediction models for never-smokers. […] Despite showing promising results, the majority of Asian risk models in our study lack external validation. […] The PLCOM2012 model demonstrated an AUC of 0.803 in the development dataset (PLCO control arm) and 0.797 in the validation set (PLCO intervention arm). […] Our meta-analysis revealed that among the eight models that underwent extensive external validations, PLCOM2012 demonstrated the best overall predictive discrimination when validated in external cohorts.
- #8 Management of lung cancer screening results based on individual prediction of current and future lung cancer riskhttps://pmc.ncbi.nlm.nih.gov/articles/PMC10186153/
We propose a risk-tailored approach for management of lung cancer screening results. This approach incorporates individual risk factors and LDCT image features into calculations of immediate and next-screen (1-year) risks of lung cancer detection, which in turn can recommend short-interval imaging or 1-year or 2-year screening intervals. […] Using individualized risk models to determine management in lung cancer screening could substantially reduce the number of screens or increase early detection. […] Compared with the NLST, the NHS England protocol would have referred 36% fewer people with abnormal screens for short-interval (e.g., 3-month) or fast-track surveillance imaging, while delaying diagnosis for only 4% of cancers. If the LCRAT+CTpos model for immediate risk had been used to identify people for short-interval or fast-track surveillance imaging, a 0.60% immediate risk-threshold would also delay diagnosis for 4% of cancers, yet would refer 42% fewer people with abnormal screens for short-interval/fast-track imaging compared with the NLST. This is a 9% additional reduction versus the NHS England protocol.
- #9 Predictive performance of risk prediction models for lung cancer incidence in Western and Asian countries: a systematic review and meta-analysis | Scientific Reportshttps://www.nature.com/articles/s41598-024-83875-6
Numerous prediction models have been developed to identify high-risk individuals for lung cancer screening, with the aim of improving early detection and survival rates. […] Extensive randomized trials, including the 2013 National Lung Screening Trial (NLST) and the 2020 NELSON trial, have demonstrated the efficacy of low-dose computed tomographic (LDCT) screening as an effective tool for early lung cancer detection. […] Our review also highlights a significant gap in prediction models for never-smokers. […] Despite showing promising results, the majority of Asian risk models in our study lack external validation. […] The PLCOM2012 model demonstrated an AUC of 0.803 in the development dataset (PLCO control arm) and 0.797 in the validation set (PLCO intervention arm). […] Our meta-analysis revealed that among the eight models that underwent extensive external validations, PLCOM2012 demonstrated the best overall predictive discrimination when validated in external cohorts.
- #10 Predictive performance of risk prediction models for lung cancer incidence in Western and Asian countries: a systematic review and meta-analysis | Scientific Reportshttps://www.nature.com/articles/s41598-024-83875-6
Numerous prediction models have been developed to identify high-risk individuals for lung cancer screening, with the aim of improving early detection and survival rates. […] Extensive randomized trials, including the 2013 National Lung Screening Trial (NLST) and the 2020 NELSON trial, have demonstrated the efficacy of low-dose computed tomographic (LDCT) screening as an effective tool for early lung cancer detection. […] Our review also highlights a significant gap in prediction models for never-smokers. […] Despite showing promising results, the majority of Asian risk models in our study lack external validation. […] The PLCOM2012 model demonstrated an AUC of 0.803 in the development dataset (PLCO control arm) and 0.797 in the validation set (PLCO intervention arm). […] Our meta-analysis revealed that among the eight models that underwent extensive external validations, PLCOM2012 demonstrated the best overall predictive discrimination when validated in external cohorts.
- #11 Six-year performance of risk-based selection for lung cancer screening in the Manchester Lung Health Check cohort | BMJ Oncologyhttps://bmjoncology.bmj.com/content/3/1/e000560
This study suggests that PLCOm2012 is an effective tool in targeted lung cancer screening. Risk models should be optimised for intended populations with differing risk profiles. Evaluation of risk-based eligibility following implementation of screening could lead to further improvements in effectiveness. […] Overall, 44% more lung cancers were diagnosed than had been expected using PLCOm2012, which predicts 6 year lung cancer risk in the absence of screening. Risk underestimation was seen in both the high-risk, screened and low-risk, unscreened groups, although cancer cases in the unscreened group were not clustered just below the eligibility threshold. […] Most lung cancers occurred in the high-risk, screened group (91%, n=105/115), and most of these were screen-detected. Three quarters of lung cancers in the high-risk cohort were early stage (76% stage I-II), in contrast to the national stage distribution prior to implementation of screening where a minority are diagnosed at early stage.
- #12 Six-year performance of risk-based selection for lung cancer screening in the Manchester Lung Health Check cohort | BMJ Oncologyhttps://bmjoncology.bmj.com/content/3/1/e000560
This study suggests that PLCOm2012 is an effective tool in targeted lung cancer screening. Risk models should be optimised for intended populations with differing risk profiles. Evaluation of risk-based eligibility following implementation of screening could lead to further improvements in effectiveness. […] Overall, 44% more lung cancers were diagnosed than had been expected using PLCOm2012, which predicts 6 year lung cancer risk in the absence of screening. Risk underestimation was seen in both the high-risk, screened and low-risk, unscreened groups, although cancer cases in the unscreened group were not clustered just below the eligibility threshold. […] Most lung cancers occurred in the high-risk, screened group (91%, n=105/115), and most of these were screen-detected. Three quarters of lung cancers in the high-risk cohort were early stage (76% stage I-II), in contrast to the national stage distribution prior to implementation of screening where a minority are diagnosed at early stage.
- #13 Management of lung cancer screening results based on individual prediction of current and future lung cancer riskhttps://pmc.ncbi.nlm.nih.gov/articles/PMC10186153/
In conclusion, using individualized risk models to decide who should be referred for short-interval imaging and to choose the length of lung cancer screening intervals could substantially increase the efficiency of lung cancer screening programs. It is likely that risk-based approaches to patient management in screening would improve cost-effectiveness, though this remains to be observed. Programs recommending biennial screening for most participants could substantially improve sensitivity by using risk models to identify high-risk individuals to be referred for annual screening.
- #14 Management of lung cancer screening results based on individual prediction of current and future lung cancer riskhttps://pmc.ncbi.nlm.nih.gov/articles/PMC10186153/
We propose a risk-tailored approach for management of lung cancer screening results. This approach incorporates individual risk factors and LDCT image features into calculations of immediate and next-screen (1-year) risks of lung cancer detection, which in turn can recommend short-interval imaging or 1-year or 2-year screening intervals. […] Using individualized risk models to determine management in lung cancer screening could substantially reduce the number of screens or increase early detection. […] Compared with the NLST, the NHS England protocol would have referred 36% fewer people with abnormal screens for short-interval (e.g., 3-month) or fast-track surveillance imaging, while delaying diagnosis for only 4% of cancers. If the LCRAT+CTpos model for immediate risk had been used to identify people for short-interval or fast-track surveillance imaging, a 0.60% immediate risk-threshold would also delay diagnosis for 4% of cancers, yet would refer 42% fewer people with abnormal screens for short-interval/fast-track imaging compared with the NLST. This is a 9% additional reduction versus the NHS England protocol.
- #15 Management of lung cancer screening results based on individual prediction of current and future lung cancer riskhttps://pmc.ncbi.nlm.nih.gov/articles/PMC10186153/
We propose a risk-tailored approach for management of lung cancer screening results. This approach incorporates individual risk factors and LDCT image features into calculations of immediate and next-screen (1-year) risks of lung cancer detection, which in turn can recommend short-interval imaging or 1-year or 2-year screening intervals. […] Using individualized risk models to determine management in lung cancer screening could substantially reduce the number of screens or increase early detection. […] Compared with the NLST, the NHS England protocol would have referred 36% fewer people with abnormal screens for short-interval (e.g., 3-month) or fast-track surveillance imaging, while delaying diagnosis for only 4% of cancers. If the LCRAT+CTpos model for immediate risk had been used to identify people for short-interval or fast-track surveillance imaging, a 0.60% immediate risk-threshold would also delay diagnosis for 4% of cancers, yet would refer 42% fewer people with abnormal screens for short-interval/fast-track imaging compared with the NLST. This is a 9% additional reduction versus the NHS England protocol.
- #16 Management of lung cancer screening results based on individual prediction of current and future lung cancer riskhttps://pmc.ncbi.nlm.nih.gov/articles/PMC10186153/
In conclusion, using individualized risk models to decide who should be referred for short-interval imaging and to choose the length of lung cancer screening intervals could substantially increase the efficiency of lung cancer screening programs. It is likely that risk-based approaches to patient management in screening would improve cost-effectiveness, though this remains to be observed. Programs recommending biennial screening for most participants could substantially improve sensitivity by using risk models to identify high-risk individuals to be referred for annual screening.
- #17 Could lung cancer prognosis be predicted from a drop of blood? – Oncology Centralhttps://www.oncology-central.com/can-lung-cancer-prognosis-be-predicted-from-a-drop-of-blood/
A blood-based biomarker panel significantly improves lung cancer risk assessment for lung cancer screening. […] a blood-based protein panel plus prediction model shows potential to lower lung cancer death by identifying individuals who need CT screening. […] The only approved screening test for lung cancer is low dose computed tomography (LDCT) and it has been previously reported that LDCT reduces lung cancer mortality by 20%.
- #18 Liquid biopsy in lung cancer: significance in diagnostics, prediction, and treatment monitoring | Molecular Cancer | Full Texthttps://molecular-cancer.biomedcentral.com/articles/10.1186/s12943-022-01505-z
Tumor markers contribute to the early diagnosis of lung cancer and play an important role in early detection and treatment, as well as precision medicine, individualized treatment, and prognosis prediction. […] Low-dose spiral CT (LDCT) is considered a standard method for the early diagnosis of lung cancer, which can significantly improve the survival rate of lung cancer patients. […] Liquid biopsy is emerging as a promising method for the identification of patients with a high risk of disease progression after curative surgery, as well as longitudinal monitoring for disease progression and therapy response. […] Liquid biopsy could help select candidate patients who would benefit from immunotherapy in several aspects. […] CTC detection has been reported to be used as an alternative method in patients with anaplastic lymphoma kinase (ALK) gene rearrangement in NSCLC when appropriate tissue biopsy samples could not be obtained.
- #19 Routine perioperative blood tests predict survival of resectable lung cancer | Scientific Reportshttps://www.nature.com/articles/s41598-023-44308-y
There is growing evidence that inflammatory, immunologic, and metabolic status is associated with cancer patients survival. Here, we built a simple algorithm to predict lung cancer outcome. Perioperative routine blood tests (RBT) of a cohort of patients with resectable primary lung cancer (LC) were analysed. Inflammatory, immunologic, and metabolic profiles were used to create a single algorithm (RBT index) predicting LC survival. […] Patients with a high RBT index had a higher 5-year mortality than low RBT patients (adjusted HR 1.93, 95% CI 1.622.31). […] We developed a simple and easily available multifunctional tool predicting short-term and long-term survival of curatively resected LC and LM. […] The correlation of socio-demographic, clinical, and inflammatory, immunological and metabolic markers with 5-year survival is shown in Table S3 (LC) and S4 (LM). In LC cohort 836 (40%) patients died within five years, compared to 529 (47%) patients in LM cohort.
- #20 Routine perioperative blood tests predict survival of resectable lung cancer | Scientific Reportshttps://www.nature.com/articles/s41598-023-44308-y
Compared with patients with a low RBT index, patients with a high RBT level had a higher risk of a worse 5-year outcome (adjusted HR 1.93, 95% CI 1.622.31), as well as patients with an intermediate level (1.34, 1.111.61). […] The present study showed that metabolic, immunologic and inflammatory biomarkers, assessed by simple routine perioperative blood examinations, can predict short- and long-term outcomes in patients with resectable LC and LM, and developed a simple and easily available tool for patients prognostication. The RBT index was able to predict 30- and 90-day mortality, as well as 5-year survival in different settings of thoracic oncology.
- #21 Routine perioperative blood tests predict survival of resectable lung cancer | Scientific Reportshttps://www.nature.com/articles/s41598-023-44308-y
Compared with patients with a low RBT index, patients with a high RBT level had a higher risk of a worse 5-year outcome (adjusted HR 1.93, 95% CI 1.622.31), as well as patients with an intermediate level (1.34, 1.111.61). […] The present study showed that metabolic, immunologic and inflammatory biomarkers, assessed by simple routine perioperative blood examinations, can predict short- and long-term outcomes in patients with resectable LC and LM, and developed a simple and easily available tool for patients prognostication. The RBT index was able to predict 30- and 90-day mortality, as well as 5-year survival in different settings of thoracic oncology.
- #22 Liquid biopsy in lung cancer: significance in diagnostics, prediction, and treatment monitoring | Molecular Cancer | Full Texthttps://molecular-cancer.biomedcentral.com/articles/10.1186/s12943-022-01505-z
Liquid biopsy opens up a new way for the early screening, diagnosis, and treatment of lung cancer, especially when tissue samples cannot be obtained. […] However, it is necessary to further develop biomarkers that are highly sensitive and specific for the early diagnosis of lung cancer. […] In conclusion, in the future, liquid biopsy technology is expected to play a greater role in the early diagnosis, accurate drug use, dynamic monitoring, and prognosis evaluation of lung cancer.
- #23 Liquid biopsy in lung cancer: significance in diagnostics, prediction, and treatment monitoring | Molecular Cancer | Full Texthttps://molecular-cancer.biomedcentral.com/articles/10.1186/s12943-022-01505-z
Tumor markers contribute to the early diagnosis of lung cancer and play an important role in early detection and treatment, as well as precision medicine, individualized treatment, and prognosis prediction. […] Low-dose spiral CT (LDCT) is considered a standard method for the early diagnosis of lung cancer, which can significantly improve the survival rate of lung cancer patients. […] Liquid biopsy is emerging as a promising method for the identification of patients with a high risk of disease progression after curative surgery, as well as longitudinal monitoring for disease progression and therapy response. […] Liquid biopsy could help select candidate patients who would benefit from immunotherapy in several aspects. […] CTC detection has been reported to be used as an alternative method in patients with anaplastic lymphoma kinase (ALK) gene rearrangement in NSCLC when appropriate tissue biopsy samples could not be obtained.
- #24 Liquid biopsy in lung cancer: significance in diagnostics, prediction, and treatment monitoring | Molecular Cancer | Full Texthttps://molecular-cancer.biomedcentral.com/articles/10.1186/s12943-022-01505-z
Tumor markers contribute to the early diagnosis of lung cancer and play an important role in early detection and treatment, as well as precision medicine, individualized treatment, and prognosis prediction. […] Low-dose spiral CT (LDCT) is considered a standard method for the early diagnosis of lung cancer, which can significantly improve the survival rate of lung cancer patients. […] Liquid biopsy is emerging as a promising method for the identification of patients with a high risk of disease progression after curative surgery, as well as longitudinal monitoring for disease progression and therapy response. […] Liquid biopsy could help select candidate patients who would benefit from immunotherapy in several aspects. […] CTC detection has been reported to be used as an alternative method in patients with anaplastic lymphoma kinase (ALK) gene rearrangement in NSCLC when appropriate tissue biopsy samples could not be obtained.
- #25 Liquid biopsy in lung cancer: significance in diagnostics, prediction, and treatment monitoring | Molecular Cancer | Full Texthttps://molecular-cancer.biomedcentral.com/articles/10.1186/s12943-022-01505-z
Therefore, it can be inferred that the DNA detected by CTCs can predict lung cancer progression and secondary drug resistance. […] Continuous monitoring of the therapeutic response of lung cancer via CTC analysis is of great value and research prospect. […] ctDNA detection is used to guide the selection of appropriate anaplastic lymphoma tyrosine kinase inhibitors (ALK-TKIs), which can provide valuable information for the precise drug treatment of lung cancer patients, especially for patients who are not suitable for biopsy. […] ctDNA analysis has shown high value in early diagnosis and prognosis. […] Therefore, gene changes, such as gene mutation, loss of heterozygosity, microsatellite instability, and gene methylation, can be found in the ctDNA of patients with lung cancer. […] The application of CTECs in the early diagnosis, efficacy detection, and prognosis prediction of lung cancer patients is still in the research stage, and more clinical validation is needed.
- #26 Liquid biopsy in lung cancer: significance in diagnostics, prediction, and treatment monitoring | Molecular Cancer | Full Texthttps://molecular-cancer.biomedcentral.com/articles/10.1186/s12943-022-01505-z
Therefore, it can be inferred that the DNA detected by CTCs can predict lung cancer progression and secondary drug resistance. […] Continuous monitoring of the therapeutic response of lung cancer via CTC analysis is of great value and research prospect. […] ctDNA detection is used to guide the selection of appropriate anaplastic lymphoma tyrosine kinase inhibitors (ALK-TKIs), which can provide valuable information for the precise drug treatment of lung cancer patients, especially for patients who are not suitable for biopsy. […] ctDNA analysis has shown high value in early diagnosis and prognosis. […] Therefore, gene changes, such as gene mutation, loss of heterozygosity, microsatellite instability, and gene methylation, can be found in the ctDNA of patients with lung cancer. […] The application of CTECs in the early diagnosis, efficacy detection, and prognosis prediction of lung cancer patients is still in the research stage, and more clinical validation is needed.
- #27 Liquid biopsy in lung cancer: significance in diagnostics, prediction, and treatment monitoring | Molecular Cancer | Full Texthttps://molecular-cancer.biomedcentral.com/articles/10.1186/s12943-022-01505-z
Therefore, it can be inferred that the DNA detected by CTCs can predict lung cancer progression and secondary drug resistance. […] Continuous monitoring of the therapeutic response of lung cancer via CTC analysis is of great value and research prospect. […] ctDNA detection is used to guide the selection of appropriate anaplastic lymphoma tyrosine kinase inhibitors (ALK-TKIs), which can provide valuable information for the precise drug treatment of lung cancer patients, especially for patients who are not suitable for biopsy. […] ctDNA analysis has shown high value in early diagnosis and prognosis. […] Therefore, gene changes, such as gene mutation, loss of heterozygosity, microsatellite instability, and gene methylation, can be found in the ctDNA of patients with lung cancer. […] The application of CTECs in the early diagnosis, efficacy detection, and prognosis prediction of lung cancer patients is still in the research stage, and more clinical validation is needed.
- #28 Liquid biopsy in lung cancer: significance in diagnostics, prediction, and treatment monitoring | Molecular Cancer | Full Texthttps://molecular-cancer.biomedcentral.com/articles/10.1186/s12943-022-01505-z
Therefore, it can be inferred that the DNA detected by CTCs can predict lung cancer progression and secondary drug resistance. […] Continuous monitoring of the therapeutic response of lung cancer via CTC analysis is of great value and research prospect. […] ctDNA detection is used to guide the selection of appropriate anaplastic lymphoma tyrosine kinase inhibitors (ALK-TKIs), which can provide valuable information for the precise drug treatment of lung cancer patients, especially for patients who are not suitable for biopsy. […] ctDNA analysis has shown high value in early diagnosis and prognosis. […] Therefore, gene changes, such as gene mutation, loss of heterozygosity, microsatellite instability, and gene methylation, can be found in the ctDNA of patients with lung cancer. […] The application of CTECs in the early diagnosis, efficacy detection, and prognosis prediction of lung cancer patients is still in the research stage, and more clinical validation is needed.
- #29 Deep Learning Empowers Lung Cancer Screening Based on Mobile Low-Dose Computed Tomography in Resource-Constrained Siteshttps://www.imrpress.com/journal/FBL/27/7/10.31083/j.fbl2707212/htm
Deep learning algorithm has the potential to alter the clinical workflow of lung cancer. […] A prospective lung cancer screening cohort was conducted in two rural areas of West China. A total of 12,360 participants were enrolled undergoing mobile CT vehicle and 86 patients were diagnosed with lung cancer after one year follow-up. […] Patients with early-stage lung cancer who received curative treatment would have a better prognosis substantially, compared with those with lung cancer at advanced stage. […] This deep learning system was developed to detect nodules (recall of 0.9507; FROC of 0.6470) and stratify the risk (ACC of 0.8696; macro-AUC of 0.8516) automatically. […] The novel approach in medical applications may assist clinicians to facilitate early diagnosis of lung cancer effectively, especially in resource-constrained sites.
- #30 Deep Learning Empowers Lung Cancer Screening Based on Mobile Low-Dose Computed Tomography in Resource-Constrained Siteshttps://www.imrpress.com/journal/FBL/27/7/10.31083/j.fbl2707212/htm
Deep learning algorithm has the potential to alter the clinical workflow of lung cancer. […] A prospective lung cancer screening cohort was conducted in two rural areas of West China. A total of 12,360 participants were enrolled undergoing mobile CT vehicle and 86 patients were diagnosed with lung cancer after one year follow-up. […] Patients with early-stage lung cancer who received curative treatment would have a better prognosis substantially, compared with those with lung cancer at advanced stage. […] This deep learning system was developed to detect nodules (recall of 0.9507; FROC of 0.6470) and stratify the risk (ACC of 0.8696; macro-AUC of 0.8516) automatically. […] The novel approach in medical applications may assist clinicians to facilitate early diagnosis of lung cancer effectively, especially in resource-constrained sites.
- #31 Deep Learning Empowers Lung Cancer Screening Based on Mobile Low-Dose Computed Tomography in Resource-Constrained Siteshttps://www.imrpress.com/journal/FBL/27/7/10.31083/j.fbl2707212/htm
Deep learning algorithm has the potential to alter the clinical workflow of lung cancer. […] A prospective lung cancer screening cohort was conducted in two rural areas of West China. A total of 12,360 participants were enrolled undergoing mobile CT vehicle and 86 patients were diagnosed with lung cancer after one year follow-up. […] Patients with early-stage lung cancer who received curative treatment would have a better prognosis substantially, compared with those with lung cancer at advanced stage. […] This deep learning system was developed to detect nodules (recall of 0.9507; FROC of 0.6470) and stratify the risk (ACC of 0.8696; macro-AUC of 0.8516) automatically. […] The novel approach in medical applications may assist clinicians to facilitate early diagnosis of lung cancer effectively, especially in resource-constrained sites.
- #32 Deep Learning Empowers Lung Cancer Screening Based on Mobile Low-Dose Computed Tomography in Resource-Constrained Siteshttps://www.imrpress.com/journal/FBL/27/7/10.31083/j.fbl2707212/htm
Deep learning algorithm has the potential to alter the clinical workflow of lung cancer. […] A prospective lung cancer screening cohort was conducted in two rural areas of West China. A total of 12,360 participants were enrolled undergoing mobile CT vehicle and 86 patients were diagnosed with lung cancer after one year follow-up. […] Patients with early-stage lung cancer who received curative treatment would have a better prognosis substantially, compared with those with lung cancer at advanced stage. […] This deep learning system was developed to detect nodules (recall of 0.9507; FROC of 0.6470) and stratify the risk (ACC of 0.8696; macro-AUC of 0.8516) automatically. […] The novel approach in medical applications may assist clinicians to facilitate early diagnosis of lung cancer effectively, especially in resource-constrained sites.
- #33 Developing and Validating a Lung Cancer Risk Prediction Model: A Nationwide Population-Based Studyhttps://www.mdpi.com/2072-6694/15/2/487
Lung cancer can be challenging to diagnose in the early stages, where treatment options are optimal. We aimed to develop 1-year prediction models for the individual risk of incident lung cancer for all individuals aged 40 or above living in Denmark on 1 January 2017. […] The objectives of this study were to develop and validate 1-year predictive models to estimate the individual risk of having lung cancer, incorporating both baseline risk factors and SES to help clinicians identify those at the highest risk for lung cancer for both individuals with no former cancer diagnosis, and in individuals with a previous cancer diagnosis. […] The best model for individuals with no previous cancer diagnosis within ten years incorporated both ICD-10 codes, ATC codes, number of GP and specialists contacts and procedures, and baseline sociodemographic characteristics.
- #34 Developing and Validating a Lung Cancer Risk Prediction Model: A Nationwide Population-Based Studyhttps://www.mdpi.com/2072-6694/15/2/487
Lung cancer can be challenging to diagnose in the early stages, where treatment options are optimal. We aimed to develop 1-year prediction models for the individual risk of incident lung cancer for all individuals aged 40 or above living in Denmark on 1 January 2017. […] The objectives of this study were to develop and validate 1-year predictive models to estimate the individual risk of having lung cancer, incorporating both baseline risk factors and SES to help clinicians identify those at the highest risk for lung cancer for both individuals with no former cancer diagnosis, and in individuals with a previous cancer diagnosis. […] The best model for individuals with no previous cancer diagnosis within ten years incorporated both ICD-10 codes, ATC codes, number of GP and specialists contacts and procedures, and baseline sociodemographic characteristics.
- #35 Developing and Validating a Lung Cancer Risk Prediction Model: A Nationwide Population-Based Studyhttps://www.mdpi.com/2072-6694/15/2/487
The validation of the model resulted in a reasonable predictive value with an AUC of 0.80 in the population with no previous cancer diagnosis compared to other predictive models of lung cancer. […] We developed and validated prediction models to support in identifying individuals at risk of lung cancer. The register-based predictive models demonstrate a potential to support clinicians and healthcare planners in identifying individuals at risk of lung cancer that should be referred for further investigation.
- #36 Developing and Validating a Lung Cancer Risk Prediction Model: A Nationwide Population-Based Studyhttps://www.mdpi.com/2072-6694/15/2/487
The validation of the model resulted in a reasonable predictive value with an AUC of 0.80 in the population with no previous cancer diagnosis compared to other predictive models of lung cancer. […] We developed and validated prediction models to support in identifying individuals at risk of lung cancer. The register-based predictive models demonstrate a potential to support clinicians and healthcare planners in identifying individuals at risk of lung cancer that should be referred for further investigation.
- #37 Six-year performance of risk-based selection for lung cancer screening in the Manchester Lung Health Check cohort | BMJ Oncologyhttps://bmjoncology.bmj.com/content/3/1/e000560
Risk prediction models are used to determine eligibility for targeted lung cancer screening. However, prospective data regarding model performance in this setting are limited. Here we report the performance of the PLCOm2012 risk model, which calculates 6 year lung cancer risk, in a cohort invited for lung cancer screening in a socioeconomically deprived area. […] Risk-based eligibility using PLCOm2012 successfully classified most eventual lung cancer cases in the high-risk, screened group. Prediction models generally underestimated risk in this socioeconomically deprived cohort, irrespective of screening status. The effect of screening on increasing the probability of lung cancer diagnosis should be considered when interpreting measures of prediction model performance. […] The Manchester Lung Health Check pilot programme applied the PLCOm2012 risk model to determine eligibility for screening in a socioeconomically deprived cohort. Follow-up of both the screened and unscreened groups demonstrated that PLCOm2012 stratified high risk participants into a high-risk group for screening, as few lung cancers arose in the low-risk, unscreened group. Risk models generally underestimated risk, due to a combination of population risk and the impact of the screening intervention.
- #38 Six-year performance of risk-based selection for lung cancer screening in the Manchester Lung Health Check cohort | BMJ Oncologyhttps://bmjoncology.bmj.com/content/3/1/e000560
This study suggests that PLCOm2012 is an effective tool in targeted lung cancer screening. Risk models should be optimised for intended populations with differing risk profiles. Evaluation of risk-based eligibility following implementation of screening could lead to further improvements in effectiveness. […] Overall, 44% more lung cancers were diagnosed than had been expected using PLCOm2012, which predicts 6 year lung cancer risk in the absence of screening. Risk underestimation was seen in both the high-risk, screened and low-risk, unscreened groups, although cancer cases in the unscreened group were not clustered just below the eligibility threshold. […] Most lung cancers occurred in the high-risk, screened group (91%, n=105/115), and most of these were screen-detected. Three quarters of lung cancers in the high-risk cohort were early stage (76% stage I-II), in contrast to the national stage distribution prior to implementation of screening where a minority are diagnosed at early stage.
- #39 Six-year performance of risk-based selection for lung cancer screening in the Manchester Lung Health Check cohort | BMJ Oncologyhttps://bmjoncology.bmj.com/content/3/1/e000560
This study suggests that PLCOm2012 is an effective tool in targeted lung cancer screening. Risk models should be optimised for intended populations with differing risk profiles. Evaluation of risk-based eligibility following implementation of screening could lead to further improvements in effectiveness. […] Overall, 44% more lung cancers were diagnosed than had been expected using PLCOm2012, which predicts 6 year lung cancer risk in the absence of screening. Risk underestimation was seen in both the high-risk, screened and low-risk, unscreened groups, although cancer cases in the unscreened group were not clustered just below the eligibility threshold. […] Most lung cancers occurred in the high-risk, screened group (91%, n=105/115), and most of these were screen-detected. Three quarters of lung cancers in the high-risk cohort were early stage (76% stage I-II), in contrast to the national stage distribution prior to implementation of screening where a minority are diagnosed at early stage.
- #40 Six-year performance of risk-based selection for lung cancer screening in the Manchester Lung Health Check cohort | BMJ Oncologyhttps://bmjoncology.bmj.com/content/3/1/e000560
Most risk prediction models analysed here underestimated risk (except LLPv2) with discrimination in the screened group ranging AUC 0.610.79. Modelling of USPSTF categorical eligibility criteria in this cohort suggested that fewer cancer cases may have been captured in the high-risk screened group. […] In this targeted lung cancer screening programme, risk-based eligibility using PLCOm2012 threshold 1.51% performed well to classify participants into high risk and low risk groups. Most participants who developed lung cancer had been identified as high risk at baseline and underwent LDCT screening, with no suggestion that a lowered eligibility threshold would have feasibly included more eventual cancer cases. However, most models, including PLCOm2012, underestimated risk. This is likely due to inherent elevated risk in the population targeted, and the screening intervention itself leading to increased probability of lung cancer diagnosis.
- #41 Predictive performance of risk prediction models for lung cancer incidence in Western and Asian countries: a systematic review and meta-analysis | Scientific Reportshttps://www.nature.com/articles/s41598-024-83875-6
Given the well-established association between smoking and lung cancer, the majority of the Western models in our study were developed among ever-smokers. […] Our review reveals a significant gap in research on risk prediction models for never-smokers, as only a small proportion of studies stratified participants based on smoking status, and only two Asian studies focused on developing a model exclusively among never-smokers. […] Therefore, understanding the interplay between genetic susceptibility and environmental factors holds promise in informing the development of a refined prediction model tailored to accurately assess lung cancer risk among Asian never-smoking populations. […] Ultimately, when deploying Western models within Asian populations, it is crucial to consider other well-established and prevalent risk factors to enhance the applicability and relevance of the model within Asian populations.
- #42 Predictive performance of risk prediction models for lung cancer incidence in Western and Asian countries: a systematic review and meta-analysis | Scientific Reportshttps://www.nature.com/articles/s41598-024-83875-6
Numerous prediction models have been developed to identify high-risk individuals for lung cancer screening, with the aim of improving early detection and survival rates. […] Extensive randomized trials, including the 2013 National Lung Screening Trial (NLST) and the 2020 NELSON trial, have demonstrated the efficacy of low-dose computed tomographic (LDCT) screening as an effective tool for early lung cancer detection. […] Our review also highlights a significant gap in prediction models for never-smokers. […] Despite showing promising results, the majority of Asian risk models in our study lack external validation. […] The PLCOM2012 model demonstrated an AUC of 0.803 in the development dataset (PLCO control arm) and 0.797 in the validation set (PLCO intervention arm). […] Our meta-analysis revealed that among the eight models that underwent extensive external validations, PLCOM2012 demonstrated the best overall predictive discrimination when validated in external cohorts.
- #43 Predictive performance of risk prediction models for lung cancer incidence in Western and Asian countries: a systematic review and meta-analysis | Scientific Reportshttps://www.nature.com/articles/s41598-024-83875-6
Given the well-established association between smoking and lung cancer, the majority of the Western models in our study were developed among ever-smokers. […] Our review reveals a significant gap in research on risk prediction models for never-smokers, as only a small proportion of studies stratified participants based on smoking status, and only two Asian studies focused on developing a model exclusively among never-smokers. […] Therefore, understanding the interplay between genetic susceptibility and environmental factors holds promise in informing the development of a refined prediction model tailored to accurately assess lung cancer risk among Asian never-smoking populations. […] Ultimately, when deploying Western models within Asian populations, it is crucial to consider other well-established and prevalent risk factors to enhance the applicability and relevance of the model within Asian populations.
- #44 Risk prediction models for lung cancer in people who have never smoked: a protocol of a systematic review | Diagnostic and Prognostic Research | Full Texthttps://diagnprognres.biomedcentral.com/articles/10.1186/s41512-024-00166-4
Early detection of lung cancer in people who have never smoked is an important public health priority. […] Currently, people who have never smoked are ineligible for lung cancer screening outside of some East Asian countries. […] With the increasing proportion of never-smokers among lung cancer cases, there is a pressing need to develop methods to identify high-risk individuals among the people who have never smoked and include them in lung cancer screening programs. […] A recent narrative review identified four risk prediction models for lung cancer in people who have never smoked. […] Consequently, there are concerns about the generalizability of these models to other settings or populations. […] Our systematic review is intended to provide a comprehensive summary of the evidence of all existing risk prediction models for lung cancer in people who have never smoked, regardless of whether they have been externally validated or not. […] Identifying and applying the most effective prediction model will facilitate personalized risk assessment for lung cancer, helping identify high-risk people who have never smoked and facilitating the optimal implementation of lung cancer screening programs in this population.
- #45 Risk prediction models for lung cancer in people who have never smoked: a protocol of a systematic review | Diagnostic and Prognostic Research | Full Texthttps://diagnprognres.biomedcentral.com/articles/10.1186/s41512-024-00166-4
Early detection of lung cancer in people who have never smoked is an important public health priority. […] Currently, people who have never smoked are ineligible for lung cancer screening outside of some East Asian countries. […] With the increasing proportion of never-smokers among lung cancer cases, there is a pressing need to develop methods to identify high-risk individuals among the people who have never smoked and include them in lung cancer screening programs. […] A recent narrative review identified four risk prediction models for lung cancer in people who have never smoked. […] Consequently, there are concerns about the generalizability of these models to other settings or populations. […] Our systematic review is intended to provide a comprehensive summary of the evidence of all existing risk prediction models for lung cancer in people who have never smoked, regardless of whether they have been externally validated or not. […] Identifying and applying the most effective prediction model will facilitate personalized risk assessment for lung cancer, helping identify high-risk people who have never smoked and facilitating the optimal implementation of lung cancer screening programs in this population.
- #46 Risk prediction models for lung cancer in people who have never smoked: a protocol of a systematic review | Diagnostic and Prognostic Research | Full Texthttps://diagnprognres.biomedcentral.com/articles/10.1186/s41512-024-00166-4
Early detection of lung cancer in people who have never smoked is an important public health priority. […] Currently, people who have never smoked are ineligible for lung cancer screening outside of some East Asian countries. […] With the increasing proportion of never-smokers among lung cancer cases, there is a pressing need to develop methods to identify high-risk individuals among the people who have never smoked and include them in lung cancer screening programs. […] A recent narrative review identified four risk prediction models for lung cancer in people who have never smoked. […] Consequently, there are concerns about the generalizability of these models to other settings or populations. […] Our systematic review is intended to provide a comprehensive summary of the evidence of all existing risk prediction models for lung cancer in people who have never smoked, regardless of whether they have been externally validated or not. […] Identifying and applying the most effective prediction model will facilitate personalized risk assessment for lung cancer, helping identify high-risk people who have never smoked and facilitating the optimal implementation of lung cancer screening programs in this population.
- #47 A narrative review of prognosis prediction models for non-small cell lung cancer: what kind of predictors should be selected and how to improve models?https://atm.amegroups.org/article/view/81325/html
The performance of gene-related models has not obviously improved. […] Most of the prevalent models are highly biased and referring to PROBAST at the beginning of the study may be able to significantly control the bias. […] Existing models should be validated in a large external dataset to make a meaningful comparison. […] Compared to the TNM staging system, a prognosis prediction model can improve the accuracy of and guide personalized therapy through the combination of multiple prognostic factors. […] The TNM staging system, WHO-PS, and pathological classification should be incorporated into all models, and the existing models should be validated in a large external dataset to make a meaningful comparison. […] The current prognosis prediction model of NSCLC is at a high ROB, and promoting the application of PROBAST may improve this situation.
- #48 A narrative review of prognosis prediction models for non-small cell lung cancer: what kind of predictors should be selected and how to improve models?https://atm.amegroups.org/article/view/81325/html
The performance of gene-related models has not obviously improved. […] Most of the prevalent models are highly biased and referring to PROBAST at the beginning of the study may be able to significantly control the bias. […] Existing models should be validated in a large external dataset to make a meaningful comparison. […] Compared to the TNM staging system, a prognosis prediction model can improve the accuracy of and guide personalized therapy through the combination of multiple prognostic factors. […] The TNM staging system, WHO-PS, and pathological classification should be incorporated into all models, and the existing models should be validated in a large external dataset to make a meaningful comparison. […] The current prognosis prediction model of NSCLC is at a high ROB, and promoting the application of PROBAST may improve this situation.
- #49 A narrative review of prognosis prediction models for non-small cell lung cancer: what kind of predictors should be selected and how to improve models?https://atm.amegroups.org/article/view/81325/html
The performance of gene-related models has not obviously improved. […] Most of the prevalent models are highly biased and referring to PROBAST at the beginning of the study may be able to significantly control the bias. […] Existing models should be validated in a large external dataset to make a meaningful comparison. […] Compared to the TNM staging system, a prognosis prediction model can improve the accuracy of and guide personalized therapy through the combination of multiple prognostic factors. […] The TNM staging system, WHO-PS, and pathological classification should be incorporated into all models, and the existing models should be validated in a large external dataset to make a meaningful comparison. […] The current prognosis prediction model of NSCLC is at a high ROB, and promoting the application of PROBAST may improve this situation.
- #50 A narrative review of prognosis prediction models for non-small cell lung cancer: what kind of predictors should be selected and how to improve models?https://atm.amegroups.org/article/view/81325/html
The performance of gene-related models has not obviously improved. […] Most of the prevalent models are highly biased and referring to PROBAST at the beginning of the study may be able to significantly control the bias. […] Existing models should be validated in a large external dataset to make a meaningful comparison. […] Compared to the TNM staging system, a prognosis prediction model can improve the accuracy of and guide personalized therapy through the combination of multiple prognostic factors. […] The TNM staging system, WHO-PS, and pathological classification should be incorporated into all models, and the existing models should be validated in a large external dataset to make a meaningful comparison. […] The current prognosis prediction model of NSCLC is at a high ROB, and promoting the application of PROBAST may improve this situation.
- #51 A narrative review of prognosis prediction models for non-small cell lung cancer: what kind of predictors should be selected and how to improve models?https://atm.amegroups.org/article/view/81325/html
The performance of gene-related models has not obviously improved. […] Most of the prevalent models are highly biased and referring to PROBAST at the beginning of the study may be able to significantly control the bias. […] Existing models should be validated in a large external dataset to make a meaningful comparison. […] Compared to the TNM staging system, a prognosis prediction model can improve the accuracy of and guide personalized therapy through the combination of multiple prognostic factors. […] The TNM staging system, WHO-PS, and pathological classification should be incorporated into all models, and the existing models should be validated in a large external dataset to make a meaningful comparison. […] The current prognosis prediction model of NSCLC is at a high ROB, and promoting the application of PROBAST may improve this situation.
- #52 A narrative review of prognosis prediction models for non-small cell lung cancer: what kind of predictors should be selected and how to improve models?https://atm.amegroups.org/article/view/81325/html
The performance of gene-related models has not obviously improved. […] Most of the prevalent models are highly biased and referring to PROBAST at the beginning of the study may be able to significantly control the bias. […] Existing models should be validated in a large external dataset to make a meaningful comparison. […] Compared to the TNM staging system, a prognosis prediction model can improve the accuracy of and guide personalized therapy through the combination of multiple prognostic factors. […] The TNM staging system, WHO-PS, and pathological classification should be incorporated into all models, and the existing models should be validated in a large external dataset to make a meaningful comparison. […] The current prognosis prediction model of NSCLC is at a high ROB, and promoting the application of PROBAST may improve this situation.
- #53 Recommendation: Lung Cancer: Screening | United States Preventive Services Taskforcehttps://www.uspreventiveservicestaskforce.org/uspstf/recommendation/lung-cancer-screening
Lung cancer has a generally poor prognosis, with an overall 5-year survival rate of 20.5%. However, early-stage lung cancer has a better prognosis and is more amenable to treatment. […] The US Preventive Services Task Force (USPSTF) concludes with moderate certainty that annual screening for lung cancer with LDCT has a moderate net benefit in persons at high risk of lung cancer based on age, total cumulative exposure to tobacco smoke, and years since quitting smoking. […] The USPSTF recommends annual screening for lung cancer with LDCT in adults aged 50 to 80 years who have a 20 pack-year smoking history and currently smoke or have quit within the past 15 years. […] The USPSTF recommends discontinuing screening once a person has not smoked for 15 years or develops a health problem that substantially limits life expectancy or the ability or willingness to have curative lung surgery. […] The moderate net benefit of screening depends on limiting screening to persons at high risk, the accuracy of image interpretation being similar to or better than that found in clinical trials, and the resolution of most false-positive results with serial imaging rather than invasive procedures.
- #54 Recommendation: Lung Cancer: Screening | United States Preventive Services Taskforcehttps://www.uspreventiveservicestaskforce.org/uspstf/recommendation/lung-cancer-screening
Lung cancer has a generally poor prognosis, with an overall 5-year survival rate of 20.5%. However, early-stage lung cancer has a better prognosis and is more amenable to treatment. […] The US Preventive Services Task Force (USPSTF) concludes with moderate certainty that annual screening for lung cancer with LDCT has a moderate net benefit in persons at high risk of lung cancer based on age, total cumulative exposure to tobacco smoke, and years since quitting smoking. […] The USPSTF recommends annual screening for lung cancer with LDCT in adults aged 50 to 80 years who have a 20 pack-year smoking history and currently smoke or have quit within the past 15 years. […] The USPSTF recommends discontinuing screening once a person has not smoked for 15 years or develops a health problem that substantially limits life expectancy or the ability or willingness to have curative lung surgery. […] The moderate net benefit of screening depends on limiting screening to persons at high risk, the accuracy of image interpretation being similar to or better than that found in clinical trials, and the resolution of most false-positive results with serial imaging rather than invasive procedures.
- #55 Recommendation: Lung Cancer: Screening | United States Preventive Services Taskforcehttps://www.uspreventiveservicestaskforce.org/uspstf/recommendation/lung-cancer-screening
Lung cancer has a generally poor prognosis, with an overall 5-year survival rate of 20.5%. However, early-stage lung cancer has a better prognosis and is more amenable to treatment. […] The US Preventive Services Task Force (USPSTF) concludes with moderate certainty that annual screening for lung cancer with LDCT has a moderate net benefit in persons at high risk of lung cancer based on age, total cumulative exposure to tobacco smoke, and years since quitting smoking. […] The USPSTF recommends annual screening for lung cancer with LDCT in adults aged 50 to 80 years who have a 20 pack-year smoking history and currently smoke or have quit within the past 15 years. […] The USPSTF recommends discontinuing screening once a person has not smoked for 15 years or develops a health problem that substantially limits life expectancy or the ability or willingness to have curative lung surgery. […] The moderate net benefit of screening depends on limiting screening to persons at high risk, the accuracy of image interpretation being similar to or better than that found in clinical trials, and the resolution of most false-positive results with serial imaging rather than invasive procedures.
- #56 NCA – Screening for Lung Cancer with Low Dose Computed Tomography (LDCT) (CAG-00439N) – Decision Memohttps://www.cms.gov/medicare-coverage-database/view/ncacal-decision-memo.aspx?proposed=N&NCAId=274
The Centers for Medicare Medicaid Services (CMS) has determined that the evidence is sufficient to add a lung cancer screening counseling and shared decision making visit, and for appropriate beneficiaries, annual screening for lung cancer with low dose computed tomography (LDCT), as an additional preventive service benefit under the Medicare program only if all of the following criteria are met: […] The USPSTF recommends annual screening for lung cancer with low-dose computed tomography in adults aged 55 to 80 years who have a 30 pack-year smoking history and currently smoke or have quit within the past 15 years. Screening should be discontinued once a person has not smoked for 15 years or develops a health problem that substantially limits life expectancy or the ability or willingness to have curative lung surgery.
- #57 NCA – Screening for Lung Cancer with Low Dose Computed Tomography (LDCT) (CAG-00439N) – Decision Memohttps://www.cms.gov/medicare-coverage-database/view/ncacal-decision-memo.aspx?proposed=N&NCAId=274
The Centers for Medicare Medicaid Services (CMS) has determined that the evidence is sufficient to add a lung cancer screening counseling and shared decision making visit, and for appropriate beneficiaries, annual screening for lung cancer with low dose computed tomography (LDCT), as an additional preventive service benefit under the Medicare program only if all of the following criteria are met: […] The USPSTF recommends annual screening for lung cancer with low-dose computed tomography in adults aged 55 to 80 years who have a 30 pack-year smoking history and currently smoke or have quit within the past 15 years. Screening should be discontinued once a person has not smoked for 15 years or develops a health problem that substantially limits life expectancy or the ability or willingness to have curative lung surgery.
- #58 NCA – Screening for Lung Cancer with Low Dose Computed Tomography (LDCT) (CAG-00439N) – Decision Memohttps://www.cms.gov/medicare-coverage-database/view/ncacal-decision-memo.aspx?proposed=N&NCAId=274
The NLST showed that people aged 55 to 74 years with a history of heavy smoking are 20 percent less likely to die from lung cancer if they are screened with low dose helical CT than with standard screening chest x-rays. […] The NLST demonstrated benefit by enrolling a large number of high exposure patients (smoking history) to be followed for several years to detect a significant decrease (247 deaths from lung cancer per 100,000 person-years in the low-dose CT group and 309 deaths per 100,000 person-years in the radiography group; number needed to screen (NNS) to prevent one lung cancer death = 320). […] Overall, based primarily on the results of the NLST, we find that the evidence is sufficient to determine that screening for lung cancer with LDCT is reasonable and necessary for the prevention or early detection of illness or disability for the exact population studied in the NLST. […] Given the burden of lung cancer on the United States population, a suitable screening test for lung cancer has been sought for many years. Lung cancer screening has been recommended by the USPSTF with a grade B recommendation for certain individuals.
- #59 Lung cancer screening – Mayo Clinichttps://www.mayoclinic.org/tests-procedures/lung-cancer-screening/about/pac-20385024
Examples of lung cancer screening results include: If no abnormalities are discovered on your lung cancer screening test, your doctor may recommend you undergo another scan in a year. You may consider continuing annual scans until you and your doctor determine they are unlikely to offer a benefit, such as if you develop other serious health problems.
- #60 Lung cancer screening – Mayo Clinichttps://www.mayoclinic.org/tests-procedures/lung-cancer-screening/about/pac-20385024
Lung cancer screening is a process that’s used to detect the presence of lung cancer in otherwise healthy people with a high risk of lung cancer. […] The goal of lung cancer screening is to detect lung cancer at a very early stage when it’s more likely to be cured. […] Studies show lung cancer screening reduces the risk of dying of lung cancer. […] Not all medical groups agree on the age at which you may consider stopping lung cancer screening. In general, continue annual lung cancer screening until you reach a point at which you’re unlikely to benefit from screening, such as when you develop other serious health conditions that may make you too frail to undergo lung cancer treatment. […] Lung cancer screening carries several risks, such as: Finding cancer that’s too advanced to cure. Advanced lung cancers, such as those that have spread, may not respond well to treatment, so finding these cancers on a lung cancer screening test might not improve or extend your life.
- #61 Lung Cancer Screening (PDQ®) – NCIhttps://www.cancer.gov/types/lung/hp/lung-screening-pdq
Based on solid evidence, screening with chest x-ray and/or sputum cytology does not reduce mortality from lung cancer in the general population or in ever-smokers. […] Given the abundance and consistency of evidence, as well as the lack of benefit observed in the PLCO trial, it is appropriate to conclude that lung cancer screening with chest x-ray and/or sputum cytology, regardless of sex or smoking status, does not reduce lung cancer mortality.
- #62 Lung Cancer Screening (PDQ®) – NCIhttps://www.cancer.gov/types/lung/hp/lung-screening-pdq
Based on solid evidence, screening with chest x-ray and/or sputum cytology does not reduce mortality from lung cancer in the general population or in ever-smokers. […] Given the abundance and consistency of evidence, as well as the lack of benefit observed in the PLCO trial, it is appropriate to conclude that lung cancer screening with chest x-ray and/or sputum cytology, regardless of sex or smoking status, does not reduce lung cancer mortality.