Women living with HIV are at greater risk for HPV (human papillomavirus)-associated gynecologic malignancies, including cervical cancer (an AIDS-defining condition) and vulvar cancer. And as life expectancy has risen for women living with HIV thanks to effective HIV therapy, so too have their rates of other gynecologic cancers (ovarian, endometrial and breast). But are all gynecologic cancers treated equally in women living with HIV?
Cancer epidemiologist Anne F. Rositch, Ph.D., of the Johns Hopkins Bloomberg School of Public Health and her colleagues sought to characterize contemporary trends, patterns of disease, adherence to national care guidelines and barriers to treatment of gynecologic malignancies among HIV-positive female patients in an urban clinic population. Rositch provided me an overview of this study, as well her investigation of cancer treatment rates in "much-older adults" living with HIV, at AIDS 2016 in Durban, South Africa.
Thanks for speaking with us, Dr. Rositch. Can you tell us a bit about this study?
The first poster we just discussed was "big data": We could go wide, but not deep. This study is the opposite of that. We have smaller numbers, but we're able to go deeper because these are actual clinic records from the Johns Hopkins Hospital. Here, we're focusing on gynecologic cancers in HIV-positive females.
All of my research has, for the last 10 years, been in HPV-associated cancers: cervical cancer in our population of women living with HIV. Cervical cancer is extremely common in our HIV population. In this study, a roughly equal percentage of cervical cancers were identified (49%) compared to all other types combined (11% endometrial, 13% ovarian, 28% vaginal/vulvar). The rates are much, much higher than our general population. It is also readily screenable and treatable. It differs, in that way, just from other cancers, which is part of what made us want to look into other [gynecologic cancers].
We had 51 cases over about a 10-year period that were in care with us at Hopkins. These include cervical, vulvar, anal, breast, uterine and ovarian cancers. We see, in general, slightly higher rates of treatment than we did in the general population of cancer patients living with HIV. This is also a younger population; this is just who we have in our catchment area in Baltimore.
Most of our women were being treated for HIV. Also, the majority of them received at least some treatment for cancer. There were only eight women who did not receive what we called "adherent care" [as defined by National Comprehensive Cancer Network (NCCN) gynecologic cancer care guidelines].
We only had one person who did not receive any cancer treatment. We had two who had an indication of treatment, but it wasn't specified what treatment they received. Sometimes there are referrals in and out [so the patient may have received treatment elsewhere].
What were some of the obstacles to treatment that you found among the women in the study who received no treatment or did not receive adherent care?
In the worst-case scenario with these cases, three out of 51 didn't receive treatment; but, really, we think it's one out of 51. And a total of eight did not receive adherent care. We have started to see what some of the medical indications were just by going through chart notes. That part is very interesting in helping us understand what we're calling a difference or a disparity -- what those reasons are [behind treatment decisions]. For example, we have one patient who had Stage 3 ovarian cancer. She died postoperatively. That was obviously a barrier to her getting adherent care, since, after her first indication of treatment, she passed.
A Stage 3 endometrial cancer patient, again, unfortunately died after their first round of chemotherapy. This is somebody who did receive treatment but did not receive adherent care. This study is trying to get into the nuances.
Then we have people who have discontinued treatment. One was due to loss of insurance; the other one had to do with comorbidities. You really see a range of reasons: Some are personal or structural, while some of them are clinical indications against treatment.
This is a highly comorbid HIV population, so unfortunately, we have mortality as one of our competing risks. That is a way in which we know this population will be different, compared to our HIV-negative cancer patients.
I see here on your poster that 60% of women in the sample had HPV-associated cancers, while 40% were diagnosed with gynecologic cancers that were not associated with HPV. Are you able to talk a bit about any differences in treatment trends that you saw between HPV-related and non-HPV-related cancers?
We're hopefully going to be able to say even more eventually. Right now we can't really say a lot [about distinctions in treatment trends]. We have small numbers, and we don't yet have that comparison group. One future step with these data: We are going to get HIV-negative comparisons -- because these are treatment trends for our HIV-positive population, but maybe this is what it is in the general population. We have to get a comparator.
Our cervical cancer cases generally get treated. We have very few of them and our clinicians are most familiar with them. But we are seeing other cancers are being treated relatively equally. So right now we are not seeing any major differences. We're expanding these data to get more case counts.
We went into this thinking that cervical cancer would be our benchmark [for comparison with other cancers]; it's one we know the most about. Now we've decided that it's an interesting benchmark, too, but we also want data for HIV-negative [patients with these cancers] to look at those differences. We're expanding to get the cases from another sister hospital in Baltimore through the University of Maryland that will get our sample size up. I will feel more comfortable saying more once we really have the numbers.
This transcript has been lightly edited for clarity.