The right target: how survival is affected by race/ethnicity, socioeconomic status

Experts debate whether targeting racial or socioeconomic disparities would have the most impact on cancer survival.

In the United States, minorities and the poor are less likely to survive cancer than whites and the affluent. The reasons for this disparity continue to be widely researched, and if medical research were able to eliminate disparities, many lives could be saved.

“If you could create a situation where the minority population got their survival up to the level of college-educated whites, you would reduce the death rate quite dramatically,” Derek Raghavan, MD, director of the Cleveland Clinic Taussig Cancer Institute and former chairman of ASCO’s Health Disparities Advisory Group, told

HemOnc Today. “But there isn’t one specific target that needs to be addressed.”

Derek Raghavan, MD, former chairman of ASCO’s Health Disparities Advisory Group, said disparity research needs to be backed by action. Photo by Cleveland Clinic

There are many answers as to why blacks, Hispanics and the poor have worse outcomes than college-educated and wealthy whites, but all of the answers seem at least partially right — or partially wrong.

“There are studies that show that the outcomes may be due to standard of care, may be due to socioeconomic status and may be due to biological differences in the types of tumors minority women are getting. There is no one answer,” said Carol A. Parise, PhD, a senior researcher at the Sutter Institute for Medical Research in Sacramento, Calif. “It’s a complicated question; all the answers are intertwined. Some studies find there is a race and socioeconomic status disparity, some will find there’s just a socioeconomic status disparity, some will find just race and some will find no disparity at all.”

Otis W. Brawley, MD, medical director for the American Cancer Society, said the reason for this is obvious.

“Every epidemiologic study that’s done that has looked at patterns of care has shown that a substantial proportion of people who we call ‘poor’ do not get the therapies that we know work,” he said. “That goes across racial lines.”

Socioeconomic disparities

Parise was part of a team of researchers in California that studied the use of radiation therapy after breast conservation surgery. In results presented in a poster session at the 2009 Breast Cancer Symposium in San Francisco, Parise and colleagues concluded that older women, blacks and the poor were far less likely to undergo radiation to treat their cancers.

“We looked at race and socioeconomic status as two separate variables,” she said in an interview. “In the lowest socioeconomic status category, you have about a 20% lower chance of getting adjuvant radiation compared with the highest category. If you’re non-Hispanic black, you have just about the same likelihood of getting radiation.”

However, race was not as important a factor as region of California. In San Francisco, approximately 80% of patients underwent radiation therapy. In Los Angeles, easily the largest of the eight geographic regions the researchers defined in the state, only 37% of patients did so. Parise said a black woman in the highest socioeconomic quintile had a 67% chance of undergoing radiation compared with 62% for a black woman in the lowest quintile. These same black women living in San Francisco or Sacramento had more than an 80% chance of receiving radiation.

“One of our speculations as to why the results came out this way is that it is dependent on where you live and if you have access to transportation. It’s possible that care centers are following different standards of care around the region,” Parise said.

Nahida H. Gordon, MD, and colleagues found something similar in 1992. In a study of OS in 253 black women and 1,132 white women diagnosed with breast cancer from 1974 to 1985, they found that race was not a significant predictor for survival after correcting for socioeconomic status.

Brawley said results of a study by Thompson et al published in the Journal of Epidemiology and Community Health in 2001 illustrate the impact of poverty on cancer. In that study, the researchers examined data collected from 21,751 Scottish women diagnosed with breast cancer from 1978 to 1987. Patients were stratified into quintiles by economic status.

Otis W. Brawley

Researchers found a difference in five-year disease-specific survival of 8.7% between the wealthiest and poorest groups. The difference increased to 10.2% at 10 years. The differences in all-cause mortality at five and 10 years were also similar. Wealthy women had longer five-year OS (71.3% vs. 55.5%) and 10-year OS (61.3% vs. 50.2%) compared with deprived women.

Thomas and colleagues also found that affluent women were less likely to undergo mastectomy (54% vs. 64%), less likely to undergo surgery alone as treatment (7% vs. 17.1%) and more likely to receive endocrine therapy (74% vs. 64%).

“This study confirmed the adverse effect of deprivation on breast cancer specific survival in Scotland, which is equally large in women aged under and over 65 years of age,” the researchers said. “These differences in outcome between the affluent and deprived groups are substantially larger than the known benefit of adjuvant systemic therapy on survival, suggesting factors such as comorbidity, immunological competence and nutrition may be involved.”

To Brawley, the implications of these results are clear.

“There are few black women in Scotland. In Scotland, they’re starting to say, ‘What is it about poverty that causes triple-negative disease?’ They’re looking not only at what the woman ate in the year before she was diagnosed with cancer and not only her BMI, they’re looking at birth weight and her BMI 40 or 50 years ago,” Brawley said. “What was in her mother’s diet when her mother was carrying her? Does poverty in the mother reset the estrogen receptor sensitivity by the fact the impoverished woman who is pregnant is likely to have a higher-fat diet than the middle-class woman who is pregnant?”

However, in the United States, with its ethnically and racially diverse population, questions remain about whether the differences in survival rates are a function of social issues or if they are intrinsic to blacks and Hispanics.

Race/ethnicity

Lack of access to care is a problem that plagues poor women and minority women in the United States. The results of Thompson’s study suggested that triple-negative breast cancer, which is more prominent among black women in the United States, may be a factor of poverty rather than race. Gordon’s results suggested that, with identical care, minority women do as well as white women.

Despite these theories, there is also evidence that patients of different races either develop different types of cancers or react differently to some treatments.

In a review published in the Journal of Clinical Oncology in 2002, Vicky L. Shavers, PhD, MPH, and Martin L. Brown, PhD, said “racial/ethnic minorities were consistently found to have less frequently received appropriate surgical resection for lung and colorectal cancers” and “racial/ethnic disparities were also consistently found in the receipt of a cancer-directed therapy, radiation therapy after breast-conserving surgery, clinical staging, and adjuvant therapy. In many cases, racial/ethnic disparities in treatment were not explained by differences in clinical profiles.”

In July, Kathy S. Albain, MD, professor of medicine at Loyola University Medical Center in Maywood, Ill., and colleagues from the Southwest Oncology Group published results showing that black patients with the same stages of breast, prostate or ovarian cancer were less likely to survive than whites who were assigned identical treatment and treated by the same doctors. Albain and colleagues reviewed results from 35 consecutive phase-3 SWOG trials conducted from 1974 to 2001. The trials studied 10 types of cancer and involved more than 19,457 patients, 11.9% of whom were black.

If poverty or access to care was the only issue, Albain told HemOnc Today, there should be no disparity because all patients received the same care and researchers adjusted for factors such as poverty and obesity. And if poverty or access to care were the source of the disparity, the survival gap would not be restricted to only those three cancers.

“The only three types of cancers that showed the disparity were organ-specific malignancies that developed in hormonally-influenced and/or reproductive organs. Thus, there has to be some clue there as to why it’s only these three specific kinds of cancers,” Albain said. “None of the theories positing that outcomes are strictly a matter of access to appropriate care address why we would see disparities in these three cancers but not all the others. With a level playing field — that is, the best doctors and identical state-of-the-art treatment, and equal access to care — still finding a significant survival disparity for these three cancers indicates to us that there is something else going on requiring increased research effort and research dollars.”

Albain said the biologic make-up, both of the patient (genes that we all inherit in different patterns that activate and metabolize toxins, drugs and hormones) and of the tumor, may interact to play an as yet undiscovered role.

“Our recommendation is that, at the same time as we’re addressing the important issues of poverty and access to care, we have to accept there is something else going on in certain malignancies that may be tracking with race, ethnicity and poverty, but may not be these issues in and of themselves,” she said. “For example, it may have a lot to do with the genes that metabolize how effective drugs are, and/or genes that metabolize and create an individuals’ level of estrogen — with all of these variables interacting with tumor biology.”

In an editorial that accompanied Albain and colleagues’ research, Brawley disagreed with Albain’s conclusions and said lack of access was still the main cause of disparities. However, he also took a more expansive view of race.

“It is important to remember that race is based in societal politics,” he wrote. “It is not a scientific categorization and is a construct rejected by anthropologists. It is more scientific to think of race as a surrogate for area of geographic origin, socioeconomic status and culture.”

Albain does not necessarily disagree with Brawley’s position, but does disagree that poverty and access explain SWOG’s results.

“It’s not a political issue. Rather, we hypothesize that over the centuries of population migration, common variations in important genes dealing with cancer outcome migrate along with the population,” she said. “There might be an interaction between these genes and hormonal levels that influence cancer biology and outcome. It would be a shame, in our opinion, to just say disparities are entirely a product of poverty and access, and if you fix those problems, outcomes will be equal. Our results showed that access is not the sole answer, which is why we are continuing to research this issue proactively.”

Raghavan said time debating this topic could be better spent because people are dying while physicians are indecisive over access vs. biology.

“The most important thing is the phrase, ‘Beware analysis paralysis,’” he said. “So many groups sit around and think about disparities and try to figure out the ultimate strategy while black Americans and Latinos are dying much faster than everybody else. It really bothers me that people are just thinking about it rather than thinking and then actually doing something.”

For the past two years, the major emphasis of the Health Disparities Advisory Group has been attracting and retaining black oncologists and those who work in underserved areas. The group established a small program to reimburse student loans for oncologists working in underserved areas.

Raghavan said ASCO has targeted minority medical students, brought them to annual meetings and set them up with mentors. The group established a summer program that attached medical students to physicians and researchers studying disparities of care.

“I absolutely agree we need to expand the cadre of African-American, Latino-American, Native American and maybe Alaskan-Native people working in oncology,” he said.

None of the above

Edith Mitchell, MD, clinical professor of medicine and medical oncology, and associate director for diversity programs at the Kimmel Cancer Center at Thomas Jefferson University in Philadelphia, said attempting to separate socioeconomic status and race is not useful. These are both important entities that may impact cancer incidence and survival.

“No one criterion marks everybody,” she said. “If you’re looking at black vs. white, not every black person fits in one category. There are numerous variables that include income data. Since there is no one multivariate that covers everybody, we’ll never know the impact of income alone. There is no real answer because [there is] no really pure population.”

Edith Mitchell

Although she is not necessarily a supporter of a government-run health care system, Mitchell said insurance and access are very important for good medical care.

“It is clear that people who have adequate insurance will seek medical care. However, we should not assume that increasing the number of individuals covered by health insurance in this country will provide an automatic fix and reform the health care system,” she said.

“There are other important components that need to be addressed, such as education, access, diet and compliance, to name a few. Increasing the number of people with insurance is a big positive step, but additional work needs to be done after that to bring about true health care reform,” Mitchell said.

Raghavan said, “It’s not just one thing. You have to do a number of things. You have to educate the community. They have to know that cancer isn’t necessarily a terminal illness; early diagnosis allows us to do less treatment with better results.”

Some of those problems may go beyond medicine. Economics and the education system may also play a role. Poor access to care may be, in some communities, a function of poor roads or a weak public transit system, and a lack of screening and preventive services may stem from a lack of education about cancer.

“It’s very unfortunate that medicine has taken the hit for a lot of these problems when, in reality, it should be society in general taking the hit,” Brawley said. “A lot of these issues are less medical than they are societal in general.”

In the end, it is possible that neither race nor socioeconomic status should matter at all.

“What we really need to do is look at the disease the patient has, the type of disease, the subtype of the tumor — look at the patient without regards to their socioeconomic status, insurance or race and treat that patient according to the standard of care for their disease,” Parise said. “It’s a matter of getting neutrality, taking a very objective look at the actual type of cancer that patient has. It goes back to looking at each case without regards to anything about the patient except their disease and finding the best way to treat this particular patient.” – by Jason Harris.

For more information:

• Albain KS. J Natl Cancer Inst. 2009;doi:10.1093/jnci/djp175.
• Ananthakrishnan AN. Arch Intern Med. 2007;167:258-264.
• Brawley OW. J Natl Cancer Inst. 2009;doi:10.1093/jnci/djp185
• Freeman HP. CA Cancer J Clin. 2004;doi:10.332/canjclin.54.2.72.
• Garner EIO. Cervical Cancer: Disparities in screening, treatment, and survival. Cancer Epidemiology, Biomarkers & Prevention. 2003;12(suppl)242s-247. Available at http://cebp.aacrjournals.org/content/12/3/242s.full. Accessed Oct. 9, 2009.
• Glanz K. J Public Health. 2003;93:202-298.
• Gordon NH. Am J Epidemiol. 1992;135:609-618.
• Lengerich EJ. J Rural Health. 2005;21:39-47.
• Loggers ET. J Clin Oncol. 2009;doi:10.1200/JCO.2009.22.4733.
• Menashe I. J Natl Cancer Inst. 2009;doi:10.1093/jnci/djp176.
• Parise CA. #140. Disparities in the use of radiation therapy following breast conserving surgery in California. Presented at: the 2009 Breast Cancer Symposium; Oct. 9-11, 2009; San Francisco.
• Rhodes R. J Clin Oncol. 2009;doi:10.1200/JCO.2009.24.2206.
• Shavers VL. J Natl Cancer Inst. 2002;doi:10.1093/jnci/94.5.334.
• Thomson CS. J Epidemiol Community Health. 2001;55:308-315.
• Williams DR. Health Aff. 2005;24:325-334.
• Woods LM. Ann Oncol. 2006;17:5-19.