Do Ultraprocessed Foods Impact Survival After Cancer?

Diets heavy in ultraprocessed foods (UPFs) are associated with earlier death in cancer survivors, a new study finds — though issues with the research design suggest that the findings should be taken with a grain of salt.

The study, published on February 4 in Cancer Epidemiology, Biomarkers & Prevention, is among the latest to point to health hazards from eating too many foods full of preservatives, dyes, and other industrially made ingredients.

These so-called UPFs have been linked to an increased risk for cancer, but whether they have any relationship to long-term survival after cancer has been unclear.

In the new study, of 802 adults with a previous cancer diagnosis, those in the top third for UPF consumption had a 48% higher rate of death from any cause over 15 years than those in the bottom third. Similarly, heavier UPF consumers had a 57% higher rate of death from cancer.

Those excess risks were seen after adjustment for numerous variables, including age, physical activity, BMI, smoking status, and socioeconomic indicators.

“Clinicians should encourage a shift toward fresh, minimally processed foods, [and] away from heavily industrially processed products,” said lead author Marialaura Bonaccio, PhD, of the Research Unit of Epidemiology and Prevention at IRCCS Neuromed in Pozzilli, Italy.

Oncologists not involved in the work said the findings support what researchers have suspected.

“UPFs have been linked to increased risk of obesity, diabetes, inflammation, cardiovascular disease, and...all-cause mortality and cardiovascular mortality,” said Urvi A. Shah, MD, a myeloma specialist who conducts nutrition research at Memorial Sloan Kettering Cancer Center in New York City. “However, there was limited data on cancer-specific mortality to date until this study.”

The findings also dovetail with recommendations on cancer prevention that emphasize diets rich in plant foods and low in processed foods, particularly those loaded with sugar, starch, and fat.

The study “may make oncologists think twice before assuring patients to ‘eat whatever you want, it doesn’t really matter’ because these investigators show that it does,” said Donald I. Abrams, MD, an integrative oncologist at the UCSF Osher Center for Integrative Health.

However, Gideon Meyerowitz-Katz, PhD, an epidemiologist at the University of Wollongong in Wollongong, Australia, was not impressed by the analysis.

He pointed to several sources of potential bias and noted that the crude results actually showed that cancer survivors with the lowest UPF consumption had a higher rate of death than the heaviest consumers.

“The story of UPFs being bad is consistent with this data, but so is the story of UPFs being fine,” said Meyerowitz-Katz, who has written about prior research on the subject.

The broad questions of whether and how UPFs might be harming human health have been gaining research interest, partly because of their ubiquity. The foods reportedly make up about 60% of the typical American diet.

There’s no universal agreement on the precise definition of “ultraprocessed,” but researchers generally use the NOVA classification system, which assigns foods into one of four groups based on the level and purpose of processing. UPFs contain ingredients not found in the standard home kitchen (such as high-fructose corn syrup) and often have artificial flavors, colors, and other additives.

Examples of UPFs include the usual “junk food,” such as candy, soda, and processed meat, but many healthy-sounding products, such as flavored yogurts and plant-based milk, also qualify.

For their study, Bonaccio and her colleagues identified 802 cancer survivors from the Moli-sani cohort study (476 women and 326 men) who completed food-frequency questionnaires an average of 8 years postdiagnosis.

Using the NOVA system, the team calculated the amount of UPF in participants’ diets as both weight and energy ratios.

Over a median follow-up of nearly 15 years, there were 281 deaths. In the lowest third of UPF consumption (4.3% mean intake by weight), there were 3.3 deaths per 100 patient-years, compared with 2.4 per 100 patient-years in the highest UPF tertile (16.7% mean intake by weight). For cancer-specific deaths, those numbers were 1.5 and 1.4, respectively.

However, after adjustment for age and total energy intake, the top UPF-intake group showed significantly higher death rates. In the final model, which adjusted for > 20 variables, the hazard ratios for the highest versus lowest UPF consumption were 1.48 (95% CI, 1.07-2.03) for all-cause mortality and 1.57 (95% CI, 1.00-2.47) for cancer mortality.

To explore potential biological mechanisms, the researchers also analyzed certain biomarkers. They found that adjustment for inflammatory markers and resting heart rate at baseline attenuated the association between UPF and all-cause deaths by nearly 40%.

The authors acknowledged some limitations of their study, including its use of self-report and potential survivor bias.

But Meyerowitz-Katz found additional weak points. For one, he said the authors “downplayed” the impact of their analysis controlling for inflammation and heart rate.

“Inflammation and heart rate are both strong markers of future cancer risk,” Meyerowitz-Katz said. “In this cohort, there would be people who were already experiencing cancer recurrence, which is important to control for at baseline.”

He also highlighted a little-known but important issue in observational research called collider bias, which can create a false association between an exposure and outcome. In this study, he said, the researchers introduced “a huge potential for collider bias” by controlling for energy intake, because both UPF consumption and cancer recurrence are causally associated with energy intake.

Bonaccio called that particular critique “a fair methodological question” but defended her work.

She pointed out that study participants were long-term survivors, which reduces the chance that their calorie intake was mainly driven by active cancer or treatment side effects.

“And,” she said, “our models include a wide set of baseline covariates that capture major determinants of both mortality and dietary intake.”

For Bonaccio, the take-home message for patients remains the same: “Emphasizing simple, home-cooked meals and traditional dietary patterns might be especially beneficial during the survivorship phase.”

The two US experts agreed that overall diet quality is key, with limits on UPFs being part of that. They also noted that the average American’s diet contains substantially more UPFs than what was seen in this Italian study.

“I spend 20 minutes of my 60-minute new patient consult in integrative oncology advising patients to eat an organic, plant-based, antioxidant-rich, anti-inflammatory, real and whole-foods diet,” Abrams said.

For her part, Shah said that cancer survivors should aim to get at least 25-30 grams of dietary fiber daily. She also suggested they avoid particular types of UPF with little to no nutritional value, such as processed meats, sugar-laden beverages, and fast food.

The study received no commercial funding. Bonaccio, Abrams, and Meyerowitz-Katz reported no financial disclosures. Shah is principle investigator on the NUTRIVENTION trial and reported receiving research funding and/or personal fees from Celgene/BMS, Janssen, and Sanofi.

A version of this article first appeared on Medscape.com.

Diets heavy in ultraprocessed foods (UPFs) are associated with earlier death in cancer survivors, a new study finds — though issues with the research design suggest that the findings should be taken with a grain of salt.

The study, published on February 4 in Cancer Epidemiology, Biomarkers & Prevention, is among the latest to point to health hazards from eating too many foods full of preservatives, dyes, and other industrially made ingredients.

These so-called UPFs have been linked to an increased risk for cancer, but whether they have any relationship to long-term survival after cancer has been unclear.

In the new study, of 802 adults with a previous cancer diagnosis, those in the top third for UPF consumption had a 48% higher rate of death from any cause over 15 years than those in the bottom third. Similarly, heavier UPF consumers had a 57% higher rate of death from cancer.

Those excess risks were seen after adjustment for numerous variables, including age, physical activity, BMI, smoking status, and socioeconomic indicators.

“Clinicians should encourage a shift toward fresh, minimally processed foods, [and] away from heavily industrially processed products,” said lead author Marialaura Bonaccio, PhD, of the Research Unit of Epidemiology and Prevention at IRCCS Neuromed in Pozzilli, Italy.

Oncologists not involved in the work said the findings support what researchers have suspected.

“UPFs have been linked to increased risk of obesity, diabetes, inflammation, cardiovascular disease, and...all-cause mortality and cardiovascular mortality,” said Urvi A. Shah, MD, a myeloma specialist who conducts nutrition research at Memorial Sloan Kettering Cancer Center in New York City. “However, there was limited data on cancer-specific mortality to date until this study.”

The findings also dovetail with recommendations on cancer prevention that emphasize diets rich in plant foods and low in processed foods, particularly those loaded with sugar, starch, and fat.

The study “may make oncologists think twice before assuring patients to ‘eat whatever you want, it doesn’t really matter’ because these investigators show that it does,” said Donald I. Abrams, MD, an integrative oncologist at the UCSF Osher Center for Integrative Health.

However, Gideon Meyerowitz-Katz, PhD, an epidemiologist at the University of Wollongong in Wollongong, Australia, was not impressed by the analysis.

He pointed to several sources of potential bias and noted that the crude results actually showed that cancer survivors with the lowest UPF consumption had a higher rate of death than the heaviest consumers.

“The story of UPFs being bad is consistent with this data, but so is the story of UPFs being fine,” said Meyerowitz-Katz, who has written about prior research on the subject.

The broad questions of whether and how UPFs might be harming human health have been gaining research interest, partly because of their ubiquity. The foods reportedly make up about 60% of the typical American diet.

There’s no universal agreement on the precise definition of “ultraprocessed,” but researchers generally use the NOVA classification system, which assigns foods into one of four groups based on the level and purpose of processing. UPFs contain ingredients not found in the standard home kitchen (such as high-fructose corn syrup) and often have artificial flavors, colors, and other additives.

Examples of UPFs include the usual “junk food,” such as candy, soda, and processed meat, but many healthy-sounding products, such as flavored yogurts and plant-based milk, also qualify.

For their study, Bonaccio and her colleagues identified 802 cancer survivors from the Moli-sani cohort study (476 women and 326 men) who completed food-frequency questionnaires an average of 8 years postdiagnosis.

Using the NOVA system, the team calculated the amount of UPF in participants’ diets as both weight and energy ratios.

Over a median follow-up of nearly 15 years, there were 281 deaths. In the lowest third of UPF consumption (4.3% mean intake by weight), there were 3.3 deaths per 100 patient-years, compared with 2.4 per 100 patient-years in the highest UPF tertile (16.7% mean intake by weight). For cancer-specific deaths, those numbers were 1.5 and 1.4, respectively.

However, after adjustment for age and total energy intake, the top UPF-intake group showed significantly higher death rates. In the final model, which adjusted for > 20 variables, the hazard ratios for the highest versus lowest UPF consumption were 1.48 (95% CI, 1.07-2.03) for all-cause mortality and 1.57 (95% CI, 1.00-2.47) for cancer mortality.

To explore potential biological mechanisms, the researchers also analyzed certain biomarkers. They found that adjustment for inflammatory markers and resting heart rate at baseline attenuated the association between UPF and all-cause deaths by nearly 40%.

The authors acknowledged some limitations of their study, including its use of self-report and potential survivor bias.

But Meyerowitz-Katz found additional weak points. For one, he said the authors “downplayed” the impact of their analysis controlling for inflammation and heart rate.

“Inflammation and heart rate are both strong markers of future cancer risk,” Meyerowitz-Katz said. “In this cohort, there would be people who were already experiencing cancer recurrence, which is important to control for at baseline.”

He also highlighted a little-known but important issue in observational research called collider bias, which can create a false association between an exposure and outcome. In this study, he said, the researchers introduced “a huge potential for collider bias” by controlling for energy intake, because both UPF consumption and cancer recurrence are causally associated with energy intake.

Bonaccio called that particular critique “a fair methodological question” but defended her work.

She pointed out that study participants were long-term survivors, which reduces the chance that their calorie intake was mainly driven by active cancer or treatment side effects.

“And,” she said, “our models include a wide set of baseline covariates that capture major determinants of both mortality and dietary intake.”

For Bonaccio, the take-home message for patients remains the same: “Emphasizing simple, home-cooked meals and traditional dietary patterns might be especially beneficial during the survivorship phase.”

The two US experts agreed that overall diet quality is key, with limits on UPFs being part of that. They also noted that the average American’s diet contains substantially more UPFs than what was seen in this Italian study.

“I spend 20 minutes of my 60-minute new patient consult in integrative oncology advising patients to eat an organic, plant-based, antioxidant-rich, anti-inflammatory, real and whole-foods diet,” Abrams said.

For her part, Shah said that cancer survivors should aim to get at least 25-30 grams of dietary fiber daily. She also suggested they avoid particular types of UPF with little to no nutritional value, such as processed meats, sugar-laden beverages, and fast food.

The study received no commercial funding. Bonaccio, Abrams, and Meyerowitz-Katz reported no financial disclosures. Shah is principle investigator on the NUTRIVENTION trial and reported receiving research funding and/or personal fees from Celgene/BMS, Janssen, and Sanofi.

A version of this article first appeared on Medscape.com.

Diets heavy in ultraprocessed foods (UPFs) are associated with earlier death in cancer survivors, a new study finds — though issues with the research design suggest that the findings should be taken with a grain of salt.

The study, published on February 4 in Cancer Epidemiology, Biomarkers & Prevention, is among the latest to point to health hazards from eating too many foods full of preservatives, dyes, and other industrially made ingredients.

These so-called UPFs have been linked to an increased risk for cancer, but whether they have any relationship to long-term survival after cancer has been unclear.

In the new study, of 802 adults with a previous cancer diagnosis, those in the top third for UPF consumption had a 48% higher rate of death from any cause over 15 years than those in the bottom third. Similarly, heavier UPF consumers had a 57% higher rate of death from cancer.

Those excess risks were seen after adjustment for numerous variables, including age, physical activity, BMI, smoking status, and socioeconomic indicators.

“Clinicians should encourage a shift toward fresh, minimally processed foods, [and] away from heavily industrially processed products,” said lead author Marialaura Bonaccio, PhD, of the Research Unit of Epidemiology and Prevention at IRCCS Neuromed in Pozzilli, Italy.

Oncologists not involved in the work said the findings support what researchers have suspected.

“UPFs have been linked to increased risk of obesity, diabetes, inflammation, cardiovascular disease, and...all-cause mortality and cardiovascular mortality,” said Urvi A. Shah, MD, a myeloma specialist who conducts nutrition research at Memorial Sloan Kettering Cancer Center in New York City. “However, there was limited data on cancer-specific mortality to date until this study.”

The findings also dovetail with recommendations on cancer prevention that emphasize diets rich in plant foods and low in processed foods, particularly those loaded with sugar, starch, and fat.

The study “may make oncologists think twice before assuring patients to ‘eat whatever you want, it doesn’t really matter’ because these investigators show that it does,” said Donald I. Abrams, MD, an integrative oncologist at the UCSF Osher Center for Integrative Health.

However, Gideon Meyerowitz-Katz, PhD, an epidemiologist at the University of Wollongong in Wollongong, Australia, was not impressed by the analysis.

He pointed to several sources of potential bias and noted that the crude results actually showed that cancer survivors with the lowest UPF consumption had a higher rate of death than the heaviest consumers.

“The story of UPFs being bad is consistent with this data, but so is the story of UPFs being fine,” said Meyerowitz-Katz, who has written about prior research on the subject.

The broad questions of whether and how UPFs might be harming human health have been gaining research interest, partly because of their ubiquity. The foods reportedly make up about 60% of the typical American diet.

There’s no universal agreement on the precise definition of “ultraprocessed,” but researchers generally use the NOVA classification system, which assigns foods into one of four groups based on the level and purpose of processing. UPFs contain ingredients not found in the standard home kitchen (such as high-fructose corn syrup) and often have artificial flavors, colors, and other additives.

Examples of UPFs include the usual “junk food,” such as candy, soda, and processed meat, but many healthy-sounding products, such as flavored yogurts and plant-based milk, also qualify.

For their study, Bonaccio and her colleagues identified 802 cancer survivors from the Moli-sani cohort study (476 women and 326 men) who completed food-frequency questionnaires an average of 8 years postdiagnosis.

Using the NOVA system, the team calculated the amount of UPF in participants’ diets as both weight and energy ratios.

Over a median follow-up of nearly 15 years, there were 281 deaths. In the lowest third of UPF consumption (4.3% mean intake by weight), there were 3.3 deaths per 100 patient-years, compared with 2.4 per 100 patient-years in the highest UPF tertile (16.7% mean intake by weight). For cancer-specific deaths, those numbers were 1.5 and 1.4, respectively.

However, after adjustment for age and total energy intake, the top UPF-intake group showed significantly higher death rates. In the final model, which adjusted for > 20 variables, the hazard ratios for the highest versus lowest UPF consumption were 1.48 (95% CI, 1.07-2.03) for all-cause mortality and 1.57 (95% CI, 1.00-2.47) for cancer mortality.

To explore potential biological mechanisms, the researchers also analyzed certain biomarkers. They found that adjustment for inflammatory markers and resting heart rate at baseline attenuated the association between UPF and all-cause deaths by nearly 40%.

The authors acknowledged some limitations of their study, including its use of self-report and potential survivor bias.

But Meyerowitz-Katz found additional weak points. For one, he said the authors “downplayed” the impact of their analysis controlling for inflammation and heart rate.

“Inflammation and heart rate are both strong markers of future cancer risk,” Meyerowitz-Katz said. “In this cohort, there would be people who were already experiencing cancer recurrence, which is important to control for at baseline.”

He also highlighted a little-known but important issue in observational research called collider bias, which can create a false association between an exposure and outcome. In this study, he said, the researchers introduced “a huge potential for collider bias” by controlling for energy intake, because both UPF consumption and cancer recurrence are causally associated with energy intake.

Bonaccio called that particular critique “a fair methodological question” but defended her work.

She pointed out that study participants were long-term survivors, which reduces the chance that their calorie intake was mainly driven by active cancer or treatment side effects.

“And,” she said, “our models include a wide set of baseline covariates that capture major determinants of both mortality and dietary intake.”

For Bonaccio, the take-home message for patients remains the same: “Emphasizing simple, home-cooked meals and traditional dietary patterns might be especially beneficial during the survivorship phase.”

The two US experts agreed that overall diet quality is key, with limits on UPFs being part of that. They also noted that the average American’s diet contains substantially more UPFs than what was seen in this Italian study.

“I spend 20 minutes of my 60-minute new patient consult in integrative oncology advising patients to eat an organic, plant-based, antioxidant-rich, anti-inflammatory, real and whole-foods diet,” Abrams said.

For her part, Shah said that cancer survivors should aim to get at least 25-30 grams of dietary fiber daily. She also suggested they avoid particular types of UPF with little to no nutritional value, such as processed meats, sugar-laden beverages, and fast food.

The study received no commercial funding. Bonaccio, Abrams, and Meyerowitz-Katz reported no financial disclosures. Shah is principle investigator on the NUTRIVENTION trial and reported receiving research funding and/or personal fees from Celgene/BMS, Janssen, and Sanofi.

A version of this article first appeared on Medscape.com.