Individuals with HIV are living longer than ever due to tremendous advances in treatment.
However, as survival has increased, so has the incidence of cancer. Malignancy is a leading cause of mortality for this population.
Individuals with HIV have a substantially higher risk for certain types of cancer than the general population. Typical AIDS-defining cancers include non-Hodgkin lymphoma and Kaposi sarcoma.
“The strongest link between HIV and cancer is related to cancers caused by certain viruses,” Eric A. Engels, MD, MPH, chief of the infections and immunoepidemiology branch in the division of cancer epidemiology & genetics at the NCI, told HemOnc Today. “People with HIV infection have a weakened immune system, so they can’t control infection with these viruses. This ultimately can lead to progression of cancer.”
However, non-AIDS-defining cancers also are becoming more prevalent for this population.
Engels and colleagues estimated the total non-AIDS-defining cancer burden will increase to 5,980 cases by 2030 — up from 5,420 cases in 2010 — with prostate and lung cancers predicted to be the most common.
“HIV-infected people are at higher risk for exposure to smoking, alcohol drinking and infections from other viruses that may be transmitted through high-risk behaviors,” Engels said. “As more people with HIV live longer and can develop cancer, we need to understand how best to help them. One way would be to understand how we can improve their immune function and how it relates to their cancer.”
HemOnc Today spoke with oncologists, infectious disease specialists and guideline panel members about the increasing burden of cancer among individuals with HIV, the barriers these individuals face to appropriate care, the urgency for guidelines to advise clinical management, the importance of collaboration between oncologists and HIV specialists, and the need for increased access to oncology clinical trials.
Increased cancer risk
Earlier during the HIV epidemic, AIDS-defining cancers represented a great concern in the community due to lack of effective therapies and inadequate immune function.
“AIDS-defining cancers are mediated in part by the immune suppression associated with HIV, so they are more likely to develop these cancers,” Gita Suneja, MD, MS, associate professor in the departments of radiation oncology and global health at Duke University, told HemOnc Today.
Individuals with HIV historically have been excluded from oncology trials, creating doubt about whether best available cancer treatments are equally effective and tolerable in this patient population, according to Gita Suneja, MD, MS.
Individuals with HIV historically have been excluded from oncology trials, creating doubt about whether best available cancer treatments are equally effective and tolerable in this patient population, according to Gita Suneja, MD, MS. “We think this explains why a lot of cancer providers hesitate when offering treatments to patients with HIV,” Suneja said.
However, treatment has greatly improved over the last few decades and, today, HIV is highly manageable with antiretroviral therapy (ART). Because HIV can be controlled, people with HIV are living longer. Also, the incidence of AIDS-defining cancers has decreased.
Islam and colleagues estimated the total cancer burden among people with HIV will decrease from 8,150 cases in 2010 to 6,690 cases by 2030, driven by strong decreases in three AIDS-defining cancers — Kaposi sarcoma, non-Hodgkin lymphoma and cervical cancer — from 2,730 in 2010 to 720 in 2030.
Based on declines that occurred from 2000 to 2012, age-specific cancer incidence rates also are expected to decrease across age groups for non-Hodgkin lymphoma, cervical cancer and lung cancer, as well as for some age groups for Kaposi sarcoma, Hodgkin lymphoma and colon cancer.
In 2014, 45% of people living with HIV were aged 50 years or older compared with 27% in 2006, according to Islam and colleagues. The proportion of people living with HIV aged 65 years or older is expected to increase from 8.5% in 2010 to 21.4% in 2030.
Although AIDS-defining cancers are projected to decrease in frequency, incidence of these cancers among individuals with HIV still exceeds that of the general population, in part due to late presentation of HIV infection or unavailability of or poor adherence to ART.
However, incidence of non-AIDS-defining cancers — related mostly to aging — is expected to increase.
“Almost every cancer has a higher risk as we age,” Erin Reid, MD, hematologist at UC San Diego Health, said in an interview. “Now that people are living longer, people living with HIV are getting the same cancers as those without HIV.”
In addition to age, different factors may increase risk for a non-AIDS-defining cancer among an individual with HIV.
Crepaz and colleagues found 44% of all new HIV diagnoses in 2016 were among black individuals. In addition, fewer black patients received ART — despite an increase in overall usage from 89% in 2009 to 94% in 2013 — compared with Hispanic and non-Hispanic white individuals.
“African-American men who have sex with men have a high rate of HIV infection compared with other groups,” Paul A. Volberding, MD, MACP, professor of medicine at University of California, San Francisco, and director of the AIDS Research Institute, said in an interview. “Men who have sex with men also tend to smoke more than the general population. So, apart from whatever might be happening with HIV, their smoking causes them to have a higher rate of smoking-related cancers.”
Other factors associated with cancer risk in this population include coinfection with other viruses, such as hepatitis B or C, as well as socioeconomic and lifestyle variables that can prevent them from seeking proper care for HIV.
For instance, a report by Goedert and colleagues — published in Trends in Cancer — indicated people living with HIV may not receive recommended cancer screening as frequently as the general population due to socioeconomic factors.
“This public health problem warrants the attention of national and community health services,” the researchers wrote. “Furthermore, in resource-limited settings, cancer screening remains a challenging issue, not only in its implementation, but also in the inability to respond to positive findings.”
The reverse also is important — screening for HIV is a key component in cancer treatment. The CDC recommends all patients with cancer be tested for HIV.
However, Hwang and colleagues found that only 18.6% of 18,874 patients who received systemic therapy for cancer were tested for HIV at initiation of treatment. Prevalence of HIV positivity from test results was 1.2%, and the prevalence of newly diagnosed HIV was 0.3%. Further, the HIV testing rate appeared lower among blacks (13.7%) than whites (19.2%).
Overall, preventing cancer among individuals with HIV is dependent on the ability to modify known risk factors combined with effective cancer screening, such as with CT scans for smokers, mammograms, cervical Pap smears and colonoscopy, according to Jeffrey Schouten, MD, JD, AAHIVE, director of the Office of HIV/AIDS Network Coordination and senior staff scientist in the vaccine and infectious disease division at Fred Hutchinson Cancer Research Center.
“How to prevent cancer in people with HIV and oncogenic viral infections, and how to diagnose the resultant cancers early so that treatment outcomes are improved, are important issues that need to be addressed for patients,” Schouten said in an interview.
Cancer prognosis should not differ among individuals with or without HIV due to the availability of newer, effective treatments for both indications.
“There isn’t a huge difference between the two because we now have drugs to treat HIV that can be used along with chemotherapy and other cancer treatments,” Volberding said. “An HIV-infected person doesn’t have much of a reason not to have the same outcomes for cancer treatment as someone without HIV.”
Still, survival rates among individuals with HIV tend to be lower across all cancer types compared with individuals without HIV.
Coghill and colleagues compared age-, sex- and race-stratified mortality among individuals with HIV with or without cancer of the lung, breast, prostate, colorectum and anus, and Hodgkin lymphoma or non-Hodgkin lymphoma.
Results showed excess mortality for all cancers except prostate cancer for at least one demographic subgroup. The difference was especially apparent for men with lung cancer (white, 573 per 1,000 person-years; nonwhite, 503 per 1,000 person-years) or non-Hodgkin lymphoma (white, 236 per 1,000 person-years; nonwhite, 261 per 1,000 person-years), and women with Hodgkin lymphoma (white, 216 per 1,000 person-years; nonwhite, 136 per 1,000 person-years) and breast cancer (nonwhite, 107 per 1,000 person-years).
“[Poor survival rates] are related to multiple factors, including the effect of HIV, late diagnosis, poor access to care and providers not treating people with HIV with the same ‘aggressiveness’ as they would treat people without HIV with the same cancer,” Schouten said.
For most cancers, people living with HIV are two to three times more likely to receive no cancer treatment than people without HIV.
Disparities in care play a large role in the undertreatment of individuals with HIV and cancer, according to the National Comprehensive Cancer Network clinical practice guidelines for individuals with cancer and HIV.
According to the guidelines, cancer treatment modifications for people with HIV should not be made based solely on a patient’s HIV status, an HIV specialist should be included in the cancer care team, and an HIV specialist along with oncology and HIV pharmacists should review proposed cancer therapy to look for possible drug-drug interactions or overlapping toxicities.
“One of the goals of the guidelines is to explicitly state that patients with HIV should get the same treatment, especially if their HIV is well-controlled,” Suneja, an author on the guidelines, said.
Still, many people with HIV do not receive any cancer treatment at all due to provider, patient and health care system barriers.
A survey of 500 oncologists, conducted by Suneja and colleagues, showed a lack of guidelines and provider education led to substandard care for individuals with HIV and cancer.
Among 60% of oncologists who responded, 20% reported not being comfortable discussing cancer treatment adverse effects and 15% were uncomfortable discussing prognosis. Seventy-nine percent reported they would provide standard cancer treatment to HIV-infected patients.
Multivariable analysis indicated physicians comfortable discussing adverse effects and prognosis of patients with HIV and cancer were more likely to provide standard cancer treatment (adjusted OR = 1.52; 95% CI, 1.12-2.07).
“We learned a substantial proportion of oncologists would change their management based on HIV status,” Suneja said.
High proficiency in the management of HIV and cancer is needed to ensure patients receive proper treatment. Provider barriers include limited resources or unfamiliarity with managing another chronic disease.
“HIV doctors are very intimidated by chemotherapy — they don’t typically understand it — and the same goes for an oncologist with HIV therapy,” Volberding said.
Comorbidities related to HIV or hepatitis coinfection, or side effects of drugs used to manage the HIV, could limit the ability to manage cancer.
“In the setting of more advanced immunosuppression, for example, a person with HIV develops AIDS and immune function is highly compromised — then you are expecting more risk for infection and complications, which could play a bigger role in how aggressively you can treat cancer,” Reid said. “In addition, there are certain cancers where we know there are potential biologic differences between people who have HIV and those whose HIV is not well controlled vs. those without HIV.”
Previous research on drug-drug interactions may also explain undertreatment. Earlier during the HIV epidemic, HIV therapies often had overlapping toxicity with chemotherapy.
“The HIV drugs today are designed to be convenient, free of toxicities and free of drug-drug interactions,” Volberding said. “If you are treating a cancer, you want HIV to be suppressed, and the drugs we use now for HIV make it easy.”
Minimizing toxicities and interactions is key for this population.
“We need to figure out how to work around, predict and minimize toxicities, and determine which drug interactions will impact drug levels, which could, in turn, impact toxicity and efficacy,” Reid said.
Patient barriers also may affect treatment receipt. These include a sense of mistrust in the health care system, embarrassment due to stigmas surrounding HIV, or demographics that might influence them not to seek care.
A lack of trust in the health care system could play a role in African-Americans with HIV seeking care.
“Outside of HIV, trusting the health care system has been a big issue for African-Americans, which also happens to be a large part of the HIV-infected community,” Volberding said. “And, they have good reason. Things like the Tuskegee syphilis experiment don’t feel ancient to them, and they remember it well.”
Having two life-threatening illnesses and managing both can also be stressful.
“Understanding challenges from the patient perspective is important and can also be another way of helping to improve outcomes,” Suneja said.
“Some of the psychosocial, mental health or substance abuse issues that may contribute to someone becoming infected with HIV may at times contribute to their willingness to treat cancer or their access to that treatment,” Schouten added.
Improving the patient experience is essential, Suneja said.
“If we can get patients plugged into the health care system earlier, then we can move the needle on improving outcomes,” she said.
Need for collaboration
The NCCN recommendations state HIV treatment for patients with cancer should be maintained by an HIV specialist, in collaboration with an oncology team. If a patient has already begun ART, then it should be continued through cancer treatment, with specialists adjusting treatments as needed.
“The most important issues for collaboration are the continuation of ART during cancer therapy and adjustments as needed for drug interactions and toxicity management,” Schouten said.
The recommendations also state that oncologists should defer to HIV or infectious disease specialists for the management of HIV, with the reverse also holding true.
A letter to the editor published in The New England Journal of Medicine demonstrated how the relationship between infectious disease specialists and oncologists is crucial during cancer treatment among those with HIV, Volberding said. After a single treatment with nivolumab (Opdivo, Bristol-Myers Squibb), three patients with lymphoma caused by human T-cell leukemia virus type 1 all experienced rapid disease progression.
Cancer requires a very intensive management of patients over a short period of time, Engels said.
“There is the shock of diagnosis and trying to help them cope and get evaluated to start treatment,” Engels said. “Cancer is a complicated disease to manage, and physicians may feel it’s too complicated to deal with.”
Because of this, communication may be inadequate between specialists.
In the study by Suneja and colleagues, 45% of oncologists reported they never or rarely discussed their cancer management plan with an HIV specialist.
“Many physicians were not discussing cancer management plans with their patient’s HIV specialist,” Suneja said. “We thought this was really concerning.”
Barriers to proper communication include patient privacy, electronic medical records and physicians’ schedules.
“As an oncologist or HIV specialist, you tend to focus on what is in front of you and your own patient and specialty,” Volberding said. “It is easy enough to forget a patient with cancer might also have other complicated conditions.”
Clinical trial participation
People living with HIV and diagnosed with cancer are often excluded from clinical trials, making data on their treatment limited.
Further, lack of inclusion in clinical trials partially drives the differential in cancer treatment between those with and without HIV.
Because researchers assumed patients with HIV had poorer prognoses, they would be excluded from cancer trials where there could be potential for complications, according to Engels.
“Now, people recognize HIV is in many cases very treatable and patients can live a long time,” Engels said. “The emphasis is that HIV should be treated as another chronic illness.”
Last year, an ASCO and Friends of Cancer Research HIV working group released a joint research statement that stated eligibility criteria for clinical trials should “strive for inclusiveness” to maximize generalizability of results. The working group recommended HIV-infected individuals who are healthy with low risk for AIDS-related outcomes should be included in cancer clinical trials unless there is specific rationale for exclusion.
The NCCN recommendations also aim to reinforce the idea that people with HIV can and should be included in oncology clinical trials if they are able.
“We expect there are fewer trials that justify the exclusion of a person solely based on their HIV status because of our ability to control HIV in many patients so they have immune function that appears to be similar to that of a non-HIV-infected individual,” Reid, also an author on the NCCN guideline, said. “We are in an era where we have not only the ability to control HIV very well, but also improve immune function. It makes it harder in some cases to justify not including a patient with HIV.”
Participation in clinical trials would generate more data for the treatment of this population.
“In oncology, we use randomized clinical trials as our gold standard of evidence. Individuals with HIV have historically been excluded, so we don’t know if the best available treatments we have are equally effective and tolerable in this patient population,” Suneja said. “We think this explains why a lot of cancer providers hesitate when offering treatments to patients with HIV.”
Fear of drug-drug interactions or complications appears to be the biggest driver for why a patient is excluded from a trial. Experts maintain that possibility for negative drug-drug interactions are an appropriate reason to exclude a patient with HIV and cancer from a clinical trial.
“If you are designing a trial with a drug that would have interactions with another drug, or the specific treatment approach could be compromised if a patient does not have well-controlled infection, a patient with HIV may not be best served by enrolling on study,” Suneja said.
Still, trials can be designed with this population in mind.
“More trials could safely include patients living with HIV, especially if they work with an HIV-cancer specialist who is knowledgeable,” Reid said.
The AIDS Malignancy Consortium — an NCI-supported clinical trials group — designs and conducts clinical trials for cancers among persons living with HIV.
Multiple such studies are underway and include: the phase 3 ANCHOR study — a comparison of topical or ablative treatment with active monitoring for and the prevention of anal cancer among patients with HIV and high-grade squamous intraepithelial lesions; a phase 2 study of safety and efficacy of recombinant EphB4-HSA fusion protein among patients with Kaposi sarcoma; and a phase 1 study of ipilimumab (Yervoy, Bristol-Myers Squibb) and nivolumab among patients with advanced HIV-associated solid tumors or Hodgkin lymphoma.
“It would be really fantastic for more HIV-infected people with cancer to participate [in trials] … but it is fair to say that we need caution in using [different] drugs with HIV to make sure we are aware of potential problems,” Volberding said.
Despite a push for more inclusion in clinical trials, questions remain about whether certain drugs or therapeutic approaches are appropriate for this subgroup.
Previous research showed HIV-positive patients with lymphoma may not be suitable for autologous hematopoietic stem cell transplants. However, a study by Joseph Alvarnas, MD, showed patients who meet standard criteria for transplant can be candidates and do not appear to have increased risk for complications.
The multicenter study included data from 40 patients (median age, 46.9 years) with treatable HIV-related lymphoma. All patients received chemotherapy and peritransplant ART.
After 100 days, 92.3% of patients were in complete remission, with one patient in partial remission. After a median follow-up of 2 years, 82% (95% CI, 65.9–91) of patients achieved 2-year OS.
Compared with outcomes of HIV-free controls from the Center for International Bone Marrow Transplant Research database, researchers observed no differences in OS and PFS probabilities between the groups.
“This is an important study because we need to better understand the long-term effects of HIV infection to ensure that patients are equitably treated in a way that respects their medical regimens and the biology of their HIV infection,” Alvarnas said in a press release.
In another study, Mehta and colleagues demonstrated that patients with HIV who underwent HSCT were not at increased risk for serious in-hospital complications.
Further, studies are underway to evaluate immunotherapies and checkpoint inhibitors — previously unstudied for patients with HIV — in this patient population.
“This is an area where science is evolving, and we know these drugs can have interesting side effects,” Volberding said. “We haven’t sorted all that out for HIV-infected patients yet, so this is also an instance where an oncologist and HIV specialist or doctor should put their heads together and work closely.”
The phase 1 CITN-12 study of pembrolizumab (Keytruda, Merck) among patients with a variety of recurrent or refractory cancers was expanded in December to include HIV-positive individuals.
“Our goal is to highlight to the investigator community that HIV-positive patients are a subgroup of individuals that should be allowed on trials, especially trials that include immune checkpoint inhibitors given the preliminary results that we saw with CITN-12,” Elad Sharon, MD, MPH, medical officer at NCI, told HemOnc Today earlier this year.
Previous data suggested PD-L1 is dysregulated in individuals with HIV, according to Thomas Uldrick, MD, MS, lead investigator of CITN-12 study and deputy head of global oncology and associate member of the vaccine and infectious disease division at Fred Hutchinson Cancer Research Center.
“As a scientific question, I was interested in PD-L1 dysregulation as something that may lead to carcinogenesis, but also something that could be targetable and potentially more useful in people with HIV than the general population,” Uldrick told HemOnc Today.
“Monoclonal antibodies, such as pembrolizumab, may block tumor or cancer growth in different ways by targeting certain cells,” Uldrick said. “It may also help the immune system kill cancer cells. This was a safety study meant to establish that immunotherapy is safe in people with HIV.”
As immunotherapy studies become more common, Uldrick said mechanisms of resistance will need to be addressed.
“In a rapidly changing field such as this, we hope that the CITN-12 study will provide evidence to allow other investigators to include HIV-positive patients in their studies. … This is a proof of concept that people with HIV who have CD4 counts as low as 100 can be safely treated with checkpoint inhibitors in diseases for which checkpoint inhibitors are already FDA approved.”
As cancer research evolves, the HIV population stands to benefit.
“We are learning cancer is an immune-mediated disease,” Suneja said. “The more we learn about that, the more potential exists for synergies with treatments for other immune diseases, leading to promising treatment that could be used for both HIV and cancer in the future.”