Guzy
Rokowania, prognozy i postęp choroby

Prognoza w onkologii stanowi ocenę rokowania i przewidywanej reakcji na leczenie nowotworu, opartą na analizie stopnia zaawansowania choroby, zajęciu węzłów chłonnych oraz czasie do nawrotu. Modele prognostyczne, takie jak CIRI (Ciągły Zindywidualizowany Indeks Ryzyka), wykorzystują dynamiczne dane kliniczne i molekularne, poprawiając dokładność przewidywań w porównaniu do tradycyjnych systemów klasyfikacji TNM. Wykorzystanie zaawansowanych technik, w tym uczenia maszynowego i analizy NLP, pozwala na integrację multimodalnych biomarkerów, co zwiększa moc dyskryminacyjną modeli prognostycznych w przewidywaniu przeżycia całkowitego i odpowiedzi na leczenie. Przykładowo, w chłoniaku rozlanym z dużych komórek B CIRI wykazał kalibrację z różnicą 5% między obserwowanymi a przewidywanymi wynikami przeżycia wolnego od zdarzeń w okresie 12-36 miesięcy.

Prognozy guzów – ogólne zasady

Prognoza to najlepsza ocena lekarza dotycząca tego, jak choroba nowotworowa wpłynie na pacjenta i jak zareaguje na leczenie. Przeżycie oznacza odsetek osób z chorobą, które żyją w pewnym momencie po diagnozie. Zarówno prognoza, jak i przeżycie zależą od wielu czynników.1 Dokładne przewidywanie długoterminowych wyników pozostaje wyzwaniem w opiece nad pacjentami z nowotworami, a lekarze szacują rokowanie, wykorzystując statystyki zebrane przez badaczy na przestrzeni wielu lat dotyczące osób z tym samym typem nowotworu.23

Lekarz może poinformować pacjenta o dobrym rokowaniu, jeśli statystyki sugerują, że nowotwór prawdopodobnie dobrze zareaguje na leczenie. Z drugiej strony, może przekazać informację o złym rokowaniu, jeśli nowotwór jest trudniejszy do opanowania. Niezależnie od tego, co powie lekarz, należy pamiętać, że prognoza jest edukowanym przypuszczeniem. Lekarz nie może być pewien, jak przebiegnie choroba u konkretnego pacjenta.4

Czynniki wpływające na prognozę

Na rokowanie w przypadku guzów wpływa wiele czynników. Głównym czynnikiem prognostycznym dla nowotworów jest stopień zaawansowania, który opisuje ilość nowotworu w organizmie, jego lokalizację i zasięg rozprzestrzeniania się.5 Nowotwór we wczesnym stadium ma mniejsze prawdopodobieństwo nawrotu, więc rokowanie jest korzystniejsze. Nowotwór zdiagnozowany w późniejszym stadium ma większe ryzyko nawrotu, więc rokowanie jest mniej korzystne.6

Ważnym czynnikiem prognostycznym jest również zajęcie węzłów chłonnych. Nowotwór, który rozprzestrzenił się do węzłów chłonnych, ma wyższe ryzyko nawrotu i mniej korzystne rokowanie niż nowotwór, który nie rozprzestrzenił się do węzłów chłonnych.7 Ponadto, im dłuższy okres przed nawrotem nowotworu, tym lepsze rokowanie. Jeśli nowotwór powraca po ponad 5 latach od diagnozy, wynik jest zwykle lepszy niż gdy nawraca mniej niż 2 lata po diagnozie.8

Modele predykcyjne w prognozowaniu guzów

Modele prognostyczne odgrywają kluczową rolę w przewidywaniu wyników leczenia i pomagają w podejmowaniu decyzji terapeutycznych. W porównaniu do systemu klasyfikacji TNM, model prognostyczny może poprawić dokładność i kierować spersonalizowaną terapią poprzez kombinację wielu czynników prognostycznych.9 Badania pokazują, że systemy oparte na danych z elektronicznych dokumentacji medycznych i sekwencjonowania DNA guza, dostarczają bezprecedensowej możliwości studiowania determinantów wyników leczenia nowotworów.10

Nowoczesne podejścia do predykcji wyników

Ciągły Zindywidualizowany Indeks Ryzyka (CIRI) to metoda dynamicznego określania prawdopodobieństwa wyników dla pojedynczych pacjentów, wykorzystująca predyktory ryzyka pozyskiwane w czasie. Stosując CIRI u pacjentów z chłoniakiem rozlanym z dużych komórek B, wykazano poprawę przewidywania wyników w porównaniu do konwencjonalnych modeli ryzyka.11 CIRI wykazał odpowiednią kalibrację przewidywań w całym przebiegu choroby, z 5% różnicą między obserwowanymi a przewidywanymi wynikami przy rozważaniu przeżycia wolnego od zdarzeń od 12 do 36 miesięcy.12

Modele oparte na uczeniu maszynowym wykorzystujące funkcje pochodzące z przetwarzania języka naturalnego, takie jak lokalizacje choroby, przewyższają te oparte wyłącznie na danych genomowych lub stadium zaawansowania, co potwierdzają badania walidacyjne i analizy na zewnętrznych, wieloinstytucjonalnych zbiorach danych.13 Wyniki wskazują, że biomarkery multimodalne są lepsze od stadium zaawansowania choroby do prognozowania, a modele łączące wiele strumieni danych, w tym zmienne wywodzące się z NLP (Natural Language Processing), mają lepszą moc dyskryminacyjną do przewidywania ogólnego przeżycia.14

Ocena skuteczności modeli predykcyjnych

Najczęstszymi wskaźnikami do oceny wydajności predykcyjnej modelu prognostycznego są dyskryminacja i kalibracja.15 Badania wykazały, że narzędzia takie jak NHS PREDICT v2.1 mają tendencję do przeszacowywania śmiertelności 5-letniej u pacjentów z przewidywanym ryzykiem powyżej 30% i śmiertelności 10-letniej u pacjentów z przewidywanym ryzykiem powyżej 50% w przypadku wczesnego raka piersi.16 Przeszacowanie 5-letniej śmiertelności zaobserwowano w przypadku raka ER-dodatniego, HER2-dodatniego, stopnia III i T3.17

Zastosowanie podejść uczenia maszynowego nie poprawiło przewidywania wyników w porównaniu z istniejącymi narzędziami, takimi jak PREDICT, jednak predyktory specyficzne dla określonych typów guzów powinny być badane w celu wspierania podejmowania decyzji w tym zakresie.18 Badania pokazują również, że mimo innowacyjności i różnorodności predyktorów, ważniejsze jest ustanowienie wysoce stabilnego modelu, który jest wygodny do zastosowania klinicznego.19

Biomarkery i nowe technologie w prognozach

Zmiany przedrakowe w piersi stanowią trudny problem decyzyjny – czy leczyć proaktywnie i zaakceptować skutki uboczne, czy czekać i ewentualnie później spotkać się z diagnozą raka inwazyjnego. Biomarker lub zestaw biomarkerów informujący o indywidualnym ryzyku progresji byłby korzystny dla pacjenta i opłacalny dla systemu opieki zdrowotnej.20

Obiecujące biomarkery w prognozowaniu

Jedną z takich cząsteczek jest wariant splicingowy OPN-c. OPN-c występuje również w części zmian przedrakowych, gdzie odzwierciedla podwyższone ryzyko progresji do raka w ciągu 5 lat, niezależnie od podtypu zmiany.21 Co godne uwagi, OPN-c występuje w części zmian przedrakowych, gdzie odzwierciedla wysokie ryzyko progresji do raka w ciągu 5 lat, niezależnie od podtypu zmiany.22 Prognozowanie ryzyka progresji z OPN-c może być połączone z OPN exon 4 w celu oceny perspektyw przeżycia.23

Badania pokazują również, że obfitość mikrobiologiczna guza, samodzielnie lub w połączeniu z ekspresją genów guza, może w pewnym stopniu przewidywać rokowanie nowotworowe i odpowiedź na leki.24 W czterech typach nowotworów – raku kory nadnerczy, płaskonabłonkowym raku szyjki macicy, glejakach niższego stopnia i czerniaku skóry podskórnej – obfitość mikrobiologiczna guza jest lepszym predyktorem rokowania niż same zmienne kliniczne.25 Ponadto, znaleziono pięć par nowotwór-lek, w których obfitość mikrobiologiczna guza jest bardziej predykcyjna dla odpowiedzi pacjenta na lek niż same zmienne kliniczne.26

Technologie radiomiczne i dosiomiczne

Wykorzystanie analizy radiomicznej jako nieinwazyjnego, szybkiego i efektywnego kosztowo podejścia do wyodrębnienia różnych ilościowych cech opartych na obrazach okazało się cenne dla prognozy pacjenta i modelowania przewidywania wyników.27 Kilka badań wykazało, że zastosowanie wielomodalnych cech radiomicznych opartych na fuzji z różnych metod obrazowania medycznego, takich jak CT, MRI i PET, może znacznie poprawić moc predykcyjną radiomiki dla innych typów nowotworów.28

Badania wykazały, że trójwymiarowy rozkład dawki zawiera cenne informacje silnie skorelowane z przewidywaniem ogólnego przeżycia u pacjentów z nowotworami głowy i szyi. Ponadto, fuzja (szczególnie z algorytmem LLRR) rozkładu dawki z obrazami CT może poprawić wydajność i dokładność niektórych modeli predykcyjnych.29 Przewidywanie ogólnego przeżycia w oparciu o radiomikę i dosiomikę oraz fuzję tych dwóch obrazów może być pomocne w procesie podejmowania decyzji i spersonalizowanym leczeniu.30

Kliniczne zastosowania modeli prognostycznych

Przewidywanie wyników leczenia i toksyczności jest kluczowe dla wyboru spersonalizowanych opcji dla indywidualnego pacjenta z rakiem.31 Opracowano modele przewidujące odległe niepowodzenie dla pacjentów z rakiem płuca i szyjki macicy po radioterapii, a także modele oparte na głębokim uczeniu do przewidywania toksyczności dla pacjentów z rakiem szyjki macicy po radioterapii.32

Spersonalizowane prognozy dla różnych typów nowotworów

Istnieje wiele zwalidowanych nomogramów zapewniających kompleksowy przegląd oczekiwanych wyników onkologicznych u pacjentów z różnymi rodzajami guzów:33

  • Model predykcyjny pooperacyjny, który zapewnia kompleksowy przegląd oczekiwanych wyników onkologicznych u pacjentów z rakiem komórek nerkowych34
  • Przedoperacyjny nomogram do przewidywania wolności od przerzutowego nawrotu w ciągu pierwszych 12 lat po radykalnej lub częściowej nefrektomii35
  • Pooperacyjny model predykcyjny oceniający ryzyko nawrotu u pacjentów z resekcją jasnokomórkowego raka nerkowokomórkowego36
  • Modele przewidujące przeżycie specyficzne dla nowotworu, prawdopodobieństwo pozostania wolnym od nawrotu w określonych ramach czasowych po operacji lub radioterapii37
  • Nomogramy do przewidywania przeżycia całkowitego po resekcji z intencją wyleczenia raka kory nadnerczy38

Modele prognostyczne dla guzów wątroby

Opracowano i zwalidowano model skali oceny ryzyka (RSSM) do stratyfikacji ryzyka prognostycznego po terapiach wewnątrztętniczych (IAT) w przypadku raka wątrobowokomórkowego (HCC). RSSM może dokładnie stratyfikować ryzyko prognostyczne dla pacjentów z HCC, którzy otrzymali IAT.39 Model ten został opracowany w oparciu o:40

RSSM dokładnie stratyfikuje pacjentów, którzy przeszli IAT, na trzy podgrupy o znacząco różnym skumulowanym długoterminowym OS, co może potencjalnie przynieść korzyści w personalizowaniu podejmowania decyzji i zmniejszyć przyszłe skutki uboczne i obciążenie ekonomiczne dla większej liczby pacjentów.41

Ograniczenia i przyszłe kierunki badań

Badania nad przewidywaniem wyników we wczesnym raku piersi są ograniczone.42 Narzędzia wspomagające podejmowanie decyzji skoncentrowane na potrzebach młodszych pacjentów z rakiem piersi powinny stać się priorytetem badawczym.43 Zastosowanie podejść uczenia maszynowego nie poprawiło przewidywania wyników w porównaniu z istniejącymi narzędziami, takimi jak PREDICT, ale predyktory specyficzne dla wczesnego raka piersi powinny być badane w celu wspierania podejmowania decyzji w tym kontekście.44

Wyzwania w modelowaniu prognostycznym

Większość powszechnych modeli jest obarczona wysokim poziomem błędu systematycznego (bias), a odwołanie się do PROBAST (narzędzie do oceny ryzyka błędu systematycznego modeli predykcyjnych) na początku badania może znacząco kontrolować ten błąd.45 Istniejące modele powinny być walidowane na dużym zewnętrznym zbiorze danych, aby umożliwić znaczące porównanie.46

Badania pokazują, że wydajność modeli związanych z genami nie uległa wyraźnej poprawie. W porównaniu do innowacyjności i różnorodności predyktorów, ważniejsze jest ustanowienie wysoce stabilnego modelu, który jest wygodny do zastosowania klinicznego.47 Szeroka aplikacja technologii sekwencjonowania nowej generacji dostarcza różnorodnych czynników predykcyjnych dla klinicznych modeli predykcyjnych. Jednak wydaje się, że biomarkery związane z genami nie mogą w sposób oczywisty poprawić wydajności modeli.48

Przyszłe kierunki rozwoju

Przewiduje się, że dynamiczna ocena ryzyka ułatwi medycynę spersonalizowaną i umożliwi innowacyjne paradygmaty terapeutyczne.49 CIRI zapewnia potencjalną drogę naprzód, aby pomóc w podejmowaniu decyzji klinicznych poprzez dostarczanie ilościowych szacunków prawdopodobnych wyników.50

Przyszłe badania powinny skupić się na aktualizacji istniejących modeli prognostycznych poprzez dostosowanie predyktorów w celu poprawy wydajności oraz promowaniu ich praktyki klinicznej poprzez zewnętrzną walidację.51 Opracowanie i walidacja tych modeli pomoże zidentyfikować pacjentów z HCC, którzy mogą odnieść korzyści z terapii systemowej, i pokieruje leczeniem.52

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.

  1. 10.04.2026
  2. www.leksykon.com.pl

Materiały źródłowe

  • #1 Prognosis and survival for breast cancer | Canadian Cancer Society
    https://cancer.ca/en/cancer-information/cancer-types/breast/prognosis-and-survival
    A prognosis is the doctors best estimate of how cancer will affect you and how it will respond to treatment. Survival is the percentage of people with a disease who are alive at some point in time after their diagnosis. Prognosis and survival depend on many factors. […] The stage is the main prognostic factor for breast cancer. It describes how much cancer is in the body, where it is and how far it has spread. […] Early-stage breast cancer is less likely to come back (recur), so it has a more favourable prognosis. Breast cancer that is diagnosed at a later stage has a greater risk of recurrence, so it has a less favourable prognosis. […] Breast cancer that has spread to lymph nodes has a higher risk of coming back and a less favourable prognosis than breast cancer that has not spread to any lymph nodes.
  • #2 Dynamic risk profiling using serial tumor biomarkers for personalized outcome Prediction
    https://pmc.ncbi.nlm.nih.gov/articles/PMC7380118/
    Accurate prediction of long-term outcomes remains a challenge in the care of cancer patients. […] We describe the Continuous Individualized Risk Index, a method to dynamically determine outcome probabilities for individual patients utilizing risk-predictors acquired over time. […] Applying CIRI to patients with diffuse large B-cell lymphoma, we demonstrate improved outcome prediction compared to conventional risk-models. […] We demonstrate CIRIs broader utility in analogous models of chronic lymphocytic leukemia and breast adenocarcinoma, and perform a proof-of-concept analysis demonstrating how CIRI could be used to develop predictive biomarkers for therapy selection. […] We envision that dynamic risk-assessment will facilitate personalized medicine and enable innovative therapeutic paradigms.
  • #3 Cancer Prognosis – NCI
    https://www.cancer.gov/about-cancer/diagnosis-staging/prognosis
    If you have cancer, you may have questions about how serious your cancer is and your chances of survival. The estimate of how the disease will go for you is called prognosis. […] Doctors estimate prognosis by using statistics that researchers have collected over many years about people with the same type of cancer. […] Your doctor may tell you that you have a good prognosis if statistics suggest that your cancer is likely to respond well to treatment. Or they may tell you that you have a poor prognosis if the cancer is harder to control. Whatever your doctor tells you, keep in mind that a prognosis is an educated guess. Your doctor cannot be certain how it will go for you. […] If you decide not to have treatment, the doctor who knows your situation best is in the best position to discuss your prognosis. […] Survival statistics most often come from studies that compare treatments with each other, rather than treatment with no treatment. So, it may not be easy for your doctor to give you an accurate prognosis.
  • #4 Cancer Prognosis – NCI
    https://www.cancer.gov/about-cancer/diagnosis-staging/prognosis
    If you have cancer, you may have questions about how serious your cancer is and your chances of survival. The estimate of how the disease will go for you is called prognosis. […] Doctors estimate prognosis by using statistics that researchers have collected over many years about people with the same type of cancer. […] Your doctor may tell you that you have a good prognosis if statistics suggest that your cancer is likely to respond well to treatment. Or they may tell you that you have a poor prognosis if the cancer is harder to control. Whatever your doctor tells you, keep in mind that a prognosis is an educated guess. Your doctor cannot be certain how it will go for you. […] If you decide not to have treatment, the doctor who knows your situation best is in the best position to discuss your prognosis. […] Survival statistics most often come from studies that compare treatments with each other, rather than treatment with no treatment. So, it may not be easy for your doctor to give you an accurate prognosis.
  • #5 Prognosis and survival for breast cancer | Canadian Cancer Society
    https://cancer.ca/en/cancer-information/cancer-types/breast/prognosis-and-survival
    A prognosis is the doctors best estimate of how cancer will affect you and how it will respond to treatment. Survival is the percentage of people with a disease who are alive at some point in time after their diagnosis. Prognosis and survival depend on many factors. […] The stage is the main prognostic factor for breast cancer. It describes how much cancer is in the body, where it is and how far it has spread. […] Early-stage breast cancer is less likely to come back (recur), so it has a more favourable prognosis. Breast cancer that is diagnosed at a later stage has a greater risk of recurrence, so it has a less favourable prognosis. […] Breast cancer that has spread to lymph nodes has a higher risk of coming back and a less favourable prognosis than breast cancer that has not spread to any lymph nodes.
  • #6 Prognosis and survival for breast cancer | Canadian Cancer Society
    https://cancer.ca/en/cancer-information/cancer-types/breast/prognosis-and-survival
    A prognosis is the doctors best estimate of how cancer will affect you and how it will respond to treatment. Survival is the percentage of people with a disease who are alive at some point in time after their diagnosis. Prognosis and survival depend on many factors. […] The stage is the main prognostic factor for breast cancer. It describes how much cancer is in the body, where it is and how far it has spread. […] Early-stage breast cancer is less likely to come back (recur), so it has a more favourable prognosis. Breast cancer that is diagnosed at a later stage has a greater risk of recurrence, so it has a less favourable prognosis. […] Breast cancer that has spread to lymph nodes has a higher risk of coming back and a less favourable prognosis than breast cancer that has not spread to any lymph nodes.
  • #7 Prognosis and survival for breast cancer | Canadian Cancer Society
    https://cancer.ca/en/cancer-information/cancer-types/breast/prognosis-and-survival
    A prognosis is the doctors best estimate of how cancer will affect you and how it will respond to treatment. Survival is the percentage of people with a disease who are alive at some point in time after their diagnosis. Prognosis and survival depend on many factors. […] The stage is the main prognostic factor for breast cancer. It describes how much cancer is in the body, where it is and how far it has spread. […] Early-stage breast cancer is less likely to come back (recur), so it has a more favourable prognosis. Breast cancer that is diagnosed at a later stage has a greater risk of recurrence, so it has a less favourable prognosis. […] Breast cancer that has spread to lymph nodes has a higher risk of coming back and a less favourable prognosis than breast cancer that has not spread to any lymph nodes.
  • #8 Prognosis and survival for breast cancer | Canadian Cancer Society
    https://cancer.ca/en/cancer-information/cancer-types/breast/prognosis-and-survival
    The longer the period of time before breast cancer comes back, the better the prognosis. If breast cancer comes back more than 5 years after diagnosis, the outcome is usually better than when it recurs less than 2 years after diagnosis. […] A local recurrence after lumpectomy and radiation therapy has a more favourable prognosis than cancer that comes back in other organs (called distant recurrence, or distant metastasis). […] Distant recurrence will be treated like chronic disease. This means that your healthcare team will offer treatments to slow the cancer’s spread and manage symptoms, rather than try to cure the cancer itself.
  • #9 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. […] Relative to the innovation and diversity of predictors, it is more important to establish a highly stable model that is convenient for clinical application. […] 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 involves the combination of three important factors, and most prognostic models would take this system into consideration when screening for predictors.
  • #10 Automated real-world data integration improves cancer outcome prediction | Nature
    https://www.nature.com/articles/s41586-024-08167-5
    The digitization of health records and growing availability of tumour DNA sequencing provide an opportunity to study the determinants of cancer outcomes with unprecedented richness. […] Leveraging MSK-CHORD to train machine learning models to predict overall survival, we find that models including features derived from natural language processing, such as sites of disease, outperform those based on genomic data or stage alone as tested by cross-validation and an external, multi-institution dataset. […] We demonstrate the feasibility of automated annotation from unstructured notes and its utility in predicting patient outcomes. […] The resulting data are provided as a public resource for real-world oncologic research. […] In this study, we used a large, integrated dataset to develop improved models of cancer outcome.
  • #11 Dynamic risk profiling using serial tumor biomarkers for personalized outcome Prediction
    https://pmc.ncbi.nlm.nih.gov/articles/PMC7380118/
    Accurate prediction of long-term outcomes remains a challenge in the care of cancer patients. […] We describe the Continuous Individualized Risk Index, a method to dynamically determine outcome probabilities for individual patients utilizing risk-predictors acquired over time. […] Applying CIRI to patients with diffuse large B-cell lymphoma, we demonstrate improved outcome prediction compared to conventional risk-models. […] We demonstrate CIRIs broader utility in analogous models of chronic lymphocytic leukemia and breast adenocarcinoma, and perform a proof-of-concept analysis demonstrating how CIRI could be used to develop predictive biomarkers for therapy selection. […] We envision that dynamic risk-assessment will facilitate personalized medicine and enable innovative therapeutic paradigms.
  • #12 Dynamic risk profiling using serial tumor biomarkers for personalized outcome Prediction
    https://pmc.ncbi.nlm.nih.gov/articles/PMC7380118/
    CIRI-DLBCL considers a total of six complementary risk predictors, including three established risk-factors (IPI, molecular cell of origin, and interim imaging), as well as three ctDNA risk-factors (pretreatment ctDNA levels, EMR, and MMR). […] We assessed the performance of CIRI-DLBCL in an independent validation cohort of 132 patients with available ctDNA data; the clinical characteristics of these patients are provided in Table S1. […] Importantly, prediction of EFS24 by CIRI significantly improved on the IPI when compared by C-statistic (0.81 vs 0.61; P0.001), with similar improvements over EMR, MMR, and interim PET as individual predictors. […] CIRI-DLBCL demonstrated adequate calibration of predictions throughout the disease course, with a 5% difference between observed and predicted outcomes when considering event-free survival from 12 to 36 months.
  • #13 Automated real-world data integration improves cancer outcome prediction | Nature
    https://www.nature.com/articles/s41586-024-08167-5
    The digitization of health records and growing availability of tumour DNA sequencing provide an opportunity to study the determinants of cancer outcomes with unprecedented richness. […] Leveraging MSK-CHORD to train machine learning models to predict overall survival, we find that models including features derived from natural language processing, such as sites of disease, outperform those based on genomic data or stage alone as tested by cross-validation and an external, multi-institution dataset. […] We demonstrate the feasibility of automated annotation from unstructured notes and its utility in predicting patient outcomes. […] The resulting data are provided as a public resource for real-world oncologic research. […] In this study, we used a large, integrated dataset to develop improved models of cancer outcome.
  • #14 Automated real-world data integration improves cancer outcome prediction | Nature
    https://www.nature.com/articles/s41586-024-08167-5
    Thus, our cohort allows for validation of proposed genomic-histopathologic associations. […] MSK-CHORD’s size also enables analyses of patients with less common combinations of features. […] Our results indicate that multimodal biomarkers are superior to disease stage for prognostication. […] Thus, models incorporating multiple data streams including NLP-derived variables had superior discriminative power for predicting OS. […] Our results highlight the importance of multiple data streams in predicting outcomes.
  • #15 Prognostic models for outcome prediction in patients with advanced hepatocellular carcinoma treated by systemic therapy: a systematic review and critical appraisal | BMC Cancer | Full Text
    https://bmccancer.biomedcentral.com/articles/10.1186/s12885-022-09841-5
    The most common indicators for evaluating the predictive performance of a prognostic model were discrimination and calibration. […] Thirty-two prognostic models were externally validated. […] All models had a high risk of bias, which may limit their application in clinical practice. […] The development and validation of these models will aid the identification of patients with HCC who may benefit from systemic therapy, and guide treatment. […] Future studies should focus on updating existing prognosis models by adjusting predictors to improve performance and promoting their clinical practice through external validation.
  • #16 Evaluating PREDICT and developing outcome prediction models in early-onset breast cancer using data from Alberta, Canada
    https://pmc.ncbi.nlm.nih.gov/articles/PMC12006220/
    Outcome prediction research in early-onset breast cancer (EoBC) is limited. […] This study evaluated the predictive performance of NHS PREDICT v2.1 and developed two prediction models for 5-year and 10-year all-cause mortality in a cohort of EoBC patients in Alberta, Canada. […] PREDICT v2.1 tended to overestimate 5-year mortality in those with 30% predicted risks and 10-year mortality in those with 50% predicted risks for EoBC in Alberta, Canada. […] The current study assessed the predictive performance of PREDICT’s 5-year and 10-year all-cause mortality estimates in a real-world population of patients with EoBC in Alberta, Canada. […] The calibration intercept showed that the average predicted probability was greater than the overall event proportion at 5 years, but not at 10 years. Overestimation of 5-year mortality was observed in ER-positive, HER2-positive, grade III, and T3 disease.
  • #17
    https://link.springer.com/article/10.1007/s10549-025-07654-1
    Outcome prediction research in early-onset breast cancer (EoBC) is limited. […] This study evaluated the predictive performance of NHS PREDICT v2.1 and developed two prediction models for 5-year and 10-year all-cause mortality in a cohort of EoBC patients in Alberta, Canada. […] PREDICT v2.1 tended to overestimate 5-year mortality in those with 30% predicted risks and 10-year mortality in those with 50% predicted risks for EoBC in Alberta, Canada. […] The calibration intercept showed that the average predicted probability was greater than the overall event proportion at 5 years, but not at 10 years. Overestimation of 5-year mortality was observed in ER-positive, HER2-positive, grade III, and T3 disease. […] Our study reports several calibration measures for PREDICT in the EoBC setting.
  • #18
    https://link.springer.com/article/10.1007/s10549-025-07654-1
    The other primary aim of this study was to develop and compare data-driven modeling approaches for outcome prediction in the EoBC setting. […] These findings point to the data being insufficient to support more complex data-driven approaches for outcome prediction in EoBC and may provide key insights for machine learning and clinical researchers in this setting. […] Decision-aid tools focused on the needs of younger breast cancer patients should become a research priority. […] The application of machine learning approaches did not improve outcome prediction compared with existing tools like PREDICT but predictors specific to EoBC should be investigated to support decision making in this setting.
  • #19 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. […] Relative to the innovation and diversity of predictors, it is more important to establish a highly stable model that is convenient for clinical application. […] 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 involves the combination of three important factors, and most prognostic models would take this system into consideration when screening for predictors.
  • #20 Crossroads: the role of biomarkers in the management of lumps in the breast | Oncotarget
    https://www.oncotarget.com/article/28402/text/
    Premalignant lesions in the breast pose a difficult decision-making problem, whether to treat proactively and accept the side effects or to engage in watchful waiting and possibly encounter a later diagnosis of invasive cancer. […] A biomarker or set of biomarkers to inform on the individual progression risk would be beneficial to the patient and cost-effective for the healthcare system. […] One such molecule is the OPN splice variant-c. OPN-c is also present in a fraction of the premalignant lesions, where it reflects an elevated risk for progression to cancer within 5 years, regardless of the lesions subtype. […] The prognostication of progression risk for an individual patient diagnosed with a premalignant lesion can be eminently meaningful, as it facilitates the decision whether (and how broadly) to treat preemptively or to engage in watchful waiting.
  • #21 Crossroads: the role of biomarkers in the management of lumps in the breast | Oncotarget
    https://www.oncotarget.com/article/28402/text/
    Premalignant lesions in the breast pose a difficult decision-making problem, whether to treat proactively and accept the side effects or to engage in watchful waiting and possibly encounter a later diagnosis of invasive cancer. […] A biomarker or set of biomarkers to inform on the individual progression risk would be beneficial to the patient and cost-effective for the healthcare system. […] One such molecule is the OPN splice variant-c. OPN-c is also present in a fraction of the premalignant lesions, where it reflects an elevated risk for progression to cancer within 5 years, regardless of the lesions subtype. […] The prognostication of progression risk for an individual patient diagnosed with a premalignant lesion can be eminently meaningful, as it facilitates the decision whether (and how broadly) to treat preemptively or to engage in watchful waiting.
  • #22 Crossroads: the role of biomarkers in the management of lumps in the breast | Oncotarget
    https://www.oncotarget.com/article/28402/text/
    Remarkably, OPN-c is also present in a fraction of the premalignant lesions, where it reflects a high risk for progression to cancer within 5 years, regardless of the lesions subtype. […] The prognostication of progression risk with OPN-c may be combined with OPN exon 4 to evaluate survival prospects.
  • #23 Crossroads: the role of biomarkers in the management of lumps in the breast | Oncotarget
    https://www.oncotarget.com/article/28402/text/
    Remarkably, OPN-c is also present in a fraction of the premalignant lesions, where it reflects a high risk for progression to cancer within 5 years, regardless of the lesions subtype. […] The prognostication of progression risk with OPN-c may be combined with OPN exon 4 to evaluate survival prospects.
  • #24 Predicting cancer prognosis and drug response from the tumor microbiome | Nature Communications
    https://www.nature.com/articles/s41467-022-30512-3
    Tumor gene expression is predictive of patient prognosis in some cancers. […] Here, we show that tumor microbial abundances, alone or in combination with tumor gene expression, can predict cancer prognosis and drug response to some extent microbial abundances are significantly less predictive of prognosis than gene expression, although similarly as predictive of drug response, but in mostly different cancer-drug combinations. […] We show that in four cancer types, adrenocortical carcinoma, cervical squamous cell carcinoma, brain lower grade glioma, and subcutaneous skin melanoma, tumor microbial abundances are better predictors of prognosis than clinical covariates alone. However, we find that tumor gene expression is a more powerful predictor of prognosis, across a wider range of cancer types, than microbial abundances.
  • #25 Predicting cancer prognosis and drug response from the tumor microbiome | Nature Communications
    https://www.nature.com/articles/s41467-022-30512-3
    Tumor gene expression is predictive of patient prognosis in some cancers. […] Here, we show that tumor microbial abundances, alone or in combination with tumor gene expression, can predict cancer prognosis and drug response to some extent microbial abundances are significantly less predictive of prognosis than gene expression, although similarly as predictive of drug response, but in mostly different cancer-drug combinations. […] We show that in four cancer types, adrenocortical carcinoma, cervical squamous cell carcinoma, brain lower grade glioma, and subcutaneous skin melanoma, tumor microbial abundances are better predictors of prognosis than clinical covariates alone. However, we find that tumor gene expression is a more powerful predictor of prognosis, across a wider range of cancer types, than microbial abundances.
  • #26 Predicting cancer prognosis and drug response from the tumor microbiome | Nature Communications
    https://www.nature.com/articles/s41467-022-30512-3
    Moreover, we find five cancer-drug pairs where tumor microbial abundances are more predictive of patient drug response than clinical covariates alone. […] We found six microbial abundance models that had a mean C-index score 0.6 and significantly outperformed their corresponding clinical covariate-only models. […] Overall, we found that tumor microbial abundances from Poore et al. were only marginally predictive of prognosis across the TCGA cohort, and that gene expression was a significantly more powerful predictor of prognosis. […] We believe that they can serve as leads for further experimental studies of the role of microbial species in modulating patient survival and drug response, potentially by metabolizing drug levels in the tumor microenvironment as suggested above, or by altering the immune response, either by changing the levels of specific immunometabolites or by having the tumors present specific bacterial antigens.
  • #27 Development and validation of survival prognostic models for head and neck cancer patients using machine learning and dosiomics and CT radiomics features: a multicentric study | Radiation Oncology | Full Text
    https://ro-journal.biomedcentral.com/articles/10.1186/s13014-024-02409-6
    This study aimed to investigate the value of clinical, radiomic features extracted from gross tumor volumes (GTVs) delineated on CT images, dose distributions (Dosiomics), and fusion of CT and dose distributions to predict outcomes in head and neck cancer (HNC) patients. […] Our results demonstrated that clinical features, Dosiomics and fusion of dose and CT images by specific ML-FS models could predict the overall survival of HNC patients with acceptable accuracy. […] The use of radiomics analysis as a noninvasive, fast, and cost-efficient approach to extract various image-based quantitative features has proven to be valuable for patient prognosis and outcome prediction modeling. […] Several studies have shown that using multi-modality fusion-based radiomic features from different medical imaging modalities, such as CT, MRI, and PET can significantly improve the predictive power of radiomics for other cancer types.
  • #28 Development and validation of survival prognostic models for head and neck cancer patients using machine learning and dosiomics and CT radiomics features: a multicentric study | Radiation Oncology | Full Text
    https://ro-journal.biomedcentral.com/articles/10.1186/s13014-024-02409-6
    This study aimed to investigate the value of clinical, radiomic features extracted from gross tumor volumes (GTVs) delineated on CT images, dose distributions (Dosiomics), and fusion of CT and dose distributions to predict outcomes in head and neck cancer (HNC) patients. […] Our results demonstrated that clinical features, Dosiomics and fusion of dose and CT images by specific ML-FS models could predict the overall survival of HNC patients with acceptable accuracy. […] The use of radiomics analysis as a noninvasive, fast, and cost-efficient approach to extract various image-based quantitative features has proven to be valuable for patient prognosis and outcome prediction modeling. […] Several studies have shown that using multi-modality fusion-based radiomic features from different medical imaging modalities, such as CT, MRI, and PET can significantly improve the predictive power of radiomics for other cancer types.
  • #29 Development and validation of survival prognostic models for head and neck cancer patients using machine learning and dosiomics and CT radiomics features: a multicentric study | Radiation Oncology | Full Text
    https://ro-journal.biomedcentral.com/articles/10.1186/s13014-024-02409-6
    While most studies investigated normal tissue complication prediction ability of dosiomics, few of them used radiomics and dosiomics for prognosis or outcome prediction. […] In a recent study by Cai et al., they trained a model for overall survival prediction and used different fusions of CT and dose distributions, reporting that fusion models outperformed single-modality models. […] Our results demonstrated that the 3D dose distribution included valuable information highly correlated with overall survival prediction in HNC patients. […] Moreover, fusion (especially with the LLRR algorithm) of the dose distribution with CT images can improve some prediction models performance and accuracy. […] Overall survival prediction based on radiomics and dosiomics and the fusion of these two images can be helpful in the decision-making process and personalized treatment.
  • #30 Development and validation of survival prognostic models for head and neck cancer patients using machine learning and dosiomics and CT radiomics features: a multicentric study | Radiation Oncology | Full Text
    https://ro-journal.biomedcentral.com/articles/10.1186/s13014-024-02409-6
    While most studies investigated normal tissue complication prediction ability of dosiomics, few of them used radiomics and dosiomics for prognosis or outcome prediction. […] In a recent study by Cai et al., they trained a model for overall survival prediction and used different fusions of CT and dose distributions, reporting that fusion models outperformed single-modality models. […] Our results demonstrated that the 3D dose distribution included valuable information highly correlated with overall survival prediction in HNC patients. […] Moreover, fusion (especially with the LLRR algorithm) of the dose distribution with CT images can improve some prediction models performance and accuracy. […] Overall survival prediction based on radiomics and dosiomics and the fusion of these two images can be helpful in the decision-making process and personalized treatment.
  • #31 Outcome Prediction | MAIA Lab | UT Southwestern, Dallas, Texas
    https://labs.utsouthwestern.edu/maia-lab/research/outcome-prediction
    Predicting treatment outcome and toxicity is critical to select personalized options for an individual cancer patient. […] We have developed models to predictive distant failure for lung and cervical cancer patients after radiation therapy. […] We have also developed deep learning based models to predict toxicity for cervical cancer patients after radiotherapy.
  • #32 Outcome Prediction | MAIA Lab | UT Southwestern, Dallas, Texas
    https://labs.utsouthwestern.edu/maia-lab/research/outcome-prediction
    Predicting treatment outcome and toxicity is critical to select personalized options for an individual cancer patient. […] We have developed models to predictive distant failure for lung and cervical cancer patients after radiation therapy. […] We have also developed deep learning based models to predict toxicity for cervical cancer patients after radiotherapy.
  • #33 Nomograms Local Index| Cancer Prediction Tools | Fox Chase Cancer Center
    http://www.cancernomograms.com/
    A post-operative prediction model which provides a comprehensive review of expected oncological outcomes in patients with renal cell carcinoma. […] A preoperative nomogram for predicting freedom from metastatic recurrence within the first 12 years following radical or partial nephrectomy […] A post-operative prediction model that assesses the risk of recurrence in patients with resected clear cell Renal Cell Carcinoma […] What is my cancer specific survival if I had a radical nephrectomy for clear cell cancer and I am X number of years post surgery? […] What are my chances of being alive and of being free from recurrence 1, 2, 3, 4, and 5 years after surgery? […] What is the probability of remaining cancer-free at 5 years following external beam radiation therapy […] What is the chance of remaining progression free for prostate-specific antigen recurrence after radical prostatectomy?
  • #34 Nomograms Local Index| Cancer Prediction Tools | Fox Chase Cancer Center
    http://www.cancernomograms.com/
    A post-operative prediction model which provides a comprehensive review of expected oncological outcomes in patients with renal cell carcinoma. […] A preoperative nomogram for predicting freedom from metastatic recurrence within the first 12 years following radical or partial nephrectomy […] A post-operative prediction model that assesses the risk of recurrence in patients with resected clear cell Renal Cell Carcinoma […] What is my cancer specific survival if I had a radical nephrectomy for clear cell cancer and I am X number of years post surgery? […] What are my chances of being alive and of being free from recurrence 1, 2, 3, 4, and 5 years after surgery? […] What is the probability of remaining cancer-free at 5 years following external beam radiation therapy […] What is the chance of remaining progression free for prostate-specific antigen recurrence after radical prostatectomy?
  • #35 Nomograms Local Index| Cancer Prediction Tools | Fox Chase Cancer Center
    http://www.cancernomograms.com/
    A post-operative prediction model which provides a comprehensive review of expected oncological outcomes in patients with renal cell carcinoma. […] A preoperative nomogram for predicting freedom from metastatic recurrence within the first 12 years following radical or partial nephrectomy […] A post-operative prediction model that assesses the risk of recurrence in patients with resected clear cell Renal Cell Carcinoma […] What is my cancer specific survival if I had a radical nephrectomy for clear cell cancer and I am X number of years post surgery? […] What are my chances of being alive and of being free from recurrence 1, 2, 3, 4, and 5 years after surgery? […] What is the probability of remaining cancer-free at 5 years following external beam radiation therapy […] What is the chance of remaining progression free for prostate-specific antigen recurrence after radical prostatectomy?
  • #36 Nomograms Local Index| Cancer Prediction Tools | Fox Chase Cancer Center
    http://www.cancernomograms.com/
    A post-operative prediction model which provides a comprehensive review of expected oncological outcomes in patients with renal cell carcinoma. […] A preoperative nomogram for predicting freedom from metastatic recurrence within the first 12 years following radical or partial nephrectomy […] A post-operative prediction model that assesses the risk of recurrence in patients with resected clear cell Renal Cell Carcinoma […] What is my cancer specific survival if I had a radical nephrectomy for clear cell cancer and I am X number of years post surgery? […] What are my chances of being alive and of being free from recurrence 1, 2, 3, 4, and 5 years after surgery? […] What is the probability of remaining cancer-free at 5 years following external beam radiation therapy […] What is the chance of remaining progression free for prostate-specific antigen recurrence after radical prostatectomy?
  • #37 Nomograms Local Index| Cancer Prediction Tools | Fox Chase Cancer Center
    http://www.cancernomograms.com/
    A post-operative prediction model which provides a comprehensive review of expected oncological outcomes in patients with renal cell carcinoma. […] A preoperative nomogram for predicting freedom from metastatic recurrence within the first 12 years following radical or partial nephrectomy […] A post-operative prediction model that assesses the risk of recurrence in patients with resected clear cell Renal Cell Carcinoma […] What is my cancer specific survival if I had a radical nephrectomy for clear cell cancer and I am X number of years post surgery? […] What are my chances of being alive and of being free from recurrence 1, 2, 3, 4, and 5 years after surgery? […] What is the probability of remaining cancer-free at 5 years following external beam radiation therapy […] What is the chance of remaining progression free for prostate-specific antigen recurrence after radical prostatectomy?
  • #38 Nomograms Local Index| Cancer Prediction Tools | Fox Chase Cancer Center
    http://www.cancernomograms.com/
    What is my likelihood for survival if I have platinum-based treatment for my non-seminomatous germ-cell tumor? […] Nomogram-based prediction of overall survival after regional lymph node dissection and the role of perioperative chemotherapy in penile squamous cell carcinoma: A retrospective multicenter study […] Nomograms to Predict Recurrence-Free and Overall Survival After Curative Resection of Adrenocortical Carcinoma […] What is my 3- and 5- year probability of survival, and median life expectancy if I have a radical nephroureterectomy for my upper urinary tract urothelial carcinoma?
  • #39
    https://link.springer.com/article/10.1007/s00330-024-10581-2
    To develop and validate a risk scoring scale model (RSSM) for stratifying prognostic risk after intra-arterial therapies (IATs) for hepatocellular carcinoma (HCC). […] The RSSM can stratify accurately prognostic risk for HCC patients received IAT. […] The risk scoring scale model could be easily implemented for physicians to stratify risk and predict prognosis quickly and accurately, thereby serving as a more favorable tool to strengthen individualized intra-arterial therapies and management in patients with unresectable hepatocellular carcinoma. […] The CatBoost model achieved the best discrimination when 12 top variables were input, with the AUC of 0.851 (95% confidence intervals (CI), 0.8330.868) for TD, 0.817 (95%CI, 0.7590.857) for ITD, and 0.791 (95%CI, 0.7480.834) for ETD. […] The RSSM was developed based on the immune checkpoint inhibitors (ICI) (hazard ratios (HR), 0.678; 95%CI 0.549, 0.837), tyrosine kinase inhibitors (TKI) (HR, 0.702; 95%CI 0.605, 0.814), local therapy (HR, 0.104; 95%CI 0.014, 0.747), response to the first IAT (HR, 4.221; 95%CI 2.229, 7.994), tumor size (HR, 1.054; 95%CI 1.038, 1.070), and BCLC grade (HR, 2.375; 95%CI 1.950, 2.894). […] The RSSM accurately stratifies patients who underwent IATs into three subgroups with significantly different cumulative long-term OS, which may potentially benefit personalized decision making and reduce future side effects and economic burden for more patients.
  • #40
    https://link.springer.com/article/10.1007/s00330-024-10581-2
    To develop and validate a risk scoring scale model (RSSM) for stratifying prognostic risk after intra-arterial therapies (IATs) for hepatocellular carcinoma (HCC). […] The RSSM can stratify accurately prognostic risk for HCC patients received IAT. […] The risk scoring scale model could be easily implemented for physicians to stratify risk and predict prognosis quickly and accurately, thereby serving as a more favorable tool to strengthen individualized intra-arterial therapies and management in patients with unresectable hepatocellular carcinoma. […] The CatBoost model achieved the best discrimination when 12 top variables were input, with the AUC of 0.851 (95% confidence intervals (CI), 0.8330.868) for TD, 0.817 (95%CI, 0.7590.857) for ITD, and 0.791 (95%CI, 0.7480.834) for ETD. […] The RSSM was developed based on the immune checkpoint inhibitors (ICI) (hazard ratios (HR), 0.678; 95%CI 0.549, 0.837), tyrosine kinase inhibitors (TKI) (HR, 0.702; 95%CI 0.605, 0.814), local therapy (HR, 0.104; 95%CI 0.014, 0.747), response to the first IAT (HR, 4.221; 95%CI 2.229, 7.994), tumor size (HR, 1.054; 95%CI 1.038, 1.070), and BCLC grade (HR, 2.375; 95%CI 1.950, 2.894). […] The RSSM accurately stratifies patients who underwent IATs into three subgroups with significantly different cumulative long-term OS, which may potentially benefit personalized decision making and reduce future side effects and economic burden for more patients.
  • #41
    https://link.springer.com/article/10.1007/s00330-024-10581-2
    To develop and validate a risk scoring scale model (RSSM) for stratifying prognostic risk after intra-arterial therapies (IATs) for hepatocellular carcinoma (HCC). […] The RSSM can stratify accurately prognostic risk for HCC patients received IAT. […] The risk scoring scale model could be easily implemented for physicians to stratify risk and predict prognosis quickly and accurately, thereby serving as a more favorable tool to strengthen individualized intra-arterial therapies and management in patients with unresectable hepatocellular carcinoma. […] The CatBoost model achieved the best discrimination when 12 top variables were input, with the AUC of 0.851 (95% confidence intervals (CI), 0.8330.868) for TD, 0.817 (95%CI, 0.7590.857) for ITD, and 0.791 (95%CI, 0.7480.834) for ETD. […] The RSSM was developed based on the immune checkpoint inhibitors (ICI) (hazard ratios (HR), 0.678; 95%CI 0.549, 0.837), tyrosine kinase inhibitors (TKI) (HR, 0.702; 95%CI 0.605, 0.814), local therapy (HR, 0.104; 95%CI 0.014, 0.747), response to the first IAT (HR, 4.221; 95%CI 2.229, 7.994), tumor size (HR, 1.054; 95%CI 1.038, 1.070), and BCLC grade (HR, 2.375; 95%CI 1.950, 2.894). […] The RSSM accurately stratifies patients who underwent IATs into three subgroups with significantly different cumulative long-term OS, which may potentially benefit personalized decision making and reduce future side effects and economic burden for more patients.
  • #42
    https://link.springer.com/article/10.1007/s10549-025-07654-1
    Outcome prediction research in early-onset breast cancer (EoBC) is limited. […] This study evaluated the predictive performance of NHS PREDICT v2.1 and developed two prediction models for 5-year and 10-year all-cause mortality in a cohort of EoBC patients in Alberta, Canada. […] PREDICT v2.1 tended to overestimate 5-year mortality in those with 30% predicted risks and 10-year mortality in those with 50% predicted risks for EoBC in Alberta, Canada. […] The calibration intercept showed that the average predicted probability was greater than the overall event proportion at 5 years, but not at 10 years. Overestimation of 5-year mortality was observed in ER-positive, HER2-positive, grade III, and T3 disease. […] Our study reports several calibration measures for PREDICT in the EoBC setting.
  • #43 Evaluating PREDICT and developing outcome prediction models in early-onset breast cancer using data from Alberta, Canada
    https://pmc.ncbi.nlm.nih.gov/articles/PMC12006220/
    Our findings show how missing predictors, like tumor size and Ki-67 index, influence PREDICT’s performance. […] Decision-aid tools focused on the needs of younger breast cancer patients should become a research priority. […] The application of machine learning approaches did not improve outcome prediction compared with existing tools like PREDICT but predictors specific to EoBC should be investigated to support decision making in this setting.
  • #44
    https://link.springer.com/article/10.1007/s10549-025-07654-1
    The other primary aim of this study was to develop and compare data-driven modeling approaches for outcome prediction in the EoBC setting. […] These findings point to the data being insufficient to support more complex data-driven approaches for outcome prediction in EoBC and may provide key insights for machine learning and clinical researchers in this setting. […] Decision-aid tools focused on the needs of younger breast cancer patients should become a research priority. […] The application of machine learning approaches did not improve outcome prediction compared with existing tools like PREDICT but predictors specific to EoBC should be investigated to support decision making in this setting.
  • #45 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. […] Relative to the innovation and diversity of predictors, it is more important to establish a highly stable model that is convenient for clinical application. […] 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 involves the combination of three important factors, and most prognostic models would take this system into consideration when screening for predictors.
  • #46 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. […] Relative to the innovation and diversity of predictors, it is more important to establish a highly stable model that is convenient for clinical application. […] 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 involves the combination of three important factors, and most prognostic models would take this system into consideration when screening for predictors.
  • #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. […] Relative to the innovation and diversity of predictors, it is more important to establish a highly stable model that is convenient for clinical application. […] 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 involves the combination of three important factors, and most prognostic models would take this system into consideration when screening for predictors.
  • #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
    Although some models may only include one component, such as T stage or number of positive lymph node stations, almost all research involving univariate and multi-variate analysis in the field of NSCLC incorporates the TMN system. […] The pathological classification of lung cancer has been increasingly detailed with the development and popularization of immunotherapy and targeted therapy. […] The wide application of next-generation sequencing technology provides a variety of predictive factors for clinical prediction models. However, it seems that gene-related biomarkers cannot obviously improve the performance of models. […] The current prognosis prediction model of NSCLC is at a high ROB, and promoting the application of PROBAST may improve this situation.
  • #49 Dynamic risk profiling using serial tumor biomarkers for personalized outcome Prediction
    https://pmc.ncbi.nlm.nih.gov/articles/PMC7380118/
    Accurate prediction of long-term outcomes remains a challenge in the care of cancer patients. […] We describe the Continuous Individualized Risk Index, a method to dynamically determine outcome probabilities for individual patients utilizing risk-predictors acquired over time. […] Applying CIRI to patients with diffuse large B-cell lymphoma, we demonstrate improved outcome prediction compared to conventional risk-models. […] We demonstrate CIRIs broader utility in analogous models of chronic lymphocytic leukemia and breast adenocarcinoma, and perform a proof-of-concept analysis demonstrating how CIRI could be used to develop predictive biomarkers for therapy selection. […] We envision that dynamic risk-assessment will facilitate personalized medicine and enable innovative therapeutic paradigms.
  • #50 Dynamic risk profiling using serial tumor biomarkers for personalized outcome Prediction
    https://pmc.ncbi.nlm.nih.gov/articles/PMC7380118/
    CIRI significantly stratified patients when considering either the final prediction after 3 cycles of immunochemotherapy or all risk-predictions in aggregate. […] CIRI-CLL demonstrated robust and quantitative stratification of patient outcomes. […] CIRI-BRCA significantly stratified patients with similar risk profiles for distant relapse-free survival both at the completion of therapy and throughout the course of treatment. […] CIRI provides a potential path forward to aid clinical decision-making by providing quantitative estimates of likely outcomes.
  • #51 Prognostic models for outcome prediction in patients with advanced hepatocellular carcinoma treated by systemic therapy: a systematic review and critical appraisal | BMC Cancer | Full Text
    https://bmccancer.biomedcentral.com/articles/10.1186/s12885-022-09841-5
    The most common indicators for evaluating the predictive performance of a prognostic model were discrimination and calibration. […] Thirty-two prognostic models were externally validated. […] All models had a high risk of bias, which may limit their application in clinical practice. […] The development and validation of these models will aid the identification of patients with HCC who may benefit from systemic therapy, and guide treatment. […] Future studies should focus on updating existing prognosis models by adjusting predictors to improve performance and promoting their clinical practice through external validation.
  • #52 Prognostic models for outcome prediction in patients with advanced hepatocellular carcinoma treated by systemic therapy: a systematic review and critical appraisal | BMC Cancer | Full Text
    https://bmccancer.biomedcentral.com/articles/10.1186/s12885-022-09841-5
    The most common indicators for evaluating the predictive performance of a prognostic model were discrimination and calibration. […] Thirty-two prognostic models were externally validated. […] All models had a high risk of bias, which may limit their application in clinical practice. […] The development and validation of these models will aid the identification of patients with HCC who may benefit from systemic therapy, and guide treatment. […] Future studies should focus on updating existing prognosis models by adjusting predictors to improve performance and promoting their clinical practice through external validation.