In recognition of World Cancer Day on February 4th a press release was made by National Polytechnic Institute (IPN) announcing “IPN Eliminates 100% of HPV Virus.” Shortly thereafter El Universal published their article entitled, “Mexican scientist cures the Human Papilloma Virus” which understandably got a lot of attention.
In a world where cancer is the second leading cause of death and cervical cancer is the fourth most common cancer among women according to the WHO, anything that might save us from cancer is exciting! What researcher doesn’t want to have their name indelibly marked in the annals of medical discoveries!
As I read the El Universal article I became confused at seemingly nonsensical statements about the study results of the principle investigator, Eva Ramón Gallegos. “The results of her investigation show she was able to eradicate HPV in 100% of the patients who had the virus but had no premalignant lesions, 64.3% in women with HPV and lesions, and 57.2% in women who had lesions but don’t have HPV.” …how can you eradicate HPV in patients who don’t actually have HPV?
Now, I am guessing this article was not initially written in English and perhaps there was something lost in translation, but it was enough for my skepticism to start creeping up behind me, warning me not to get too excited about this claim.
In reading the original press release, the claim is certainly misleading. It says “IPN proved the elimination of 100 percent of the Human Papilloma Virus” and leads you to believe that it cleared all 29 patients of HPV. However, several paragraphs down is the aforementioned statement which was copied in the El Universal article that says it “eradicated HPV in 100% of the patients… [without] premalignant lesions.” So the first statement is directly misleading. It doesn’t eliminate 100% of the HPV virus in all cases, only those without lesions – and it doesn’t say how many of those 29 women that included and even that number was statistically relevant.
There does seem to be some really interesting research going on, but I think it is a little lofty to say it “cures HPV” when its efficacy seems to be predicated on the presence of premalignant lesions. That’s a pretty big caveat in the world of HPV and the progression of cervical cancer. Big “Buts” are not exactly a favorable thing in the science world (sorry Sir Mix A Lot).
Another thing is a lack of published data. Usually big medical discoveries would have been published in a journal with plenty of graphs and statistics to back up their reported findings. Unfortunately, apart from IPN’s press release, I was unable to any find published data on this study involving these 29 women in Mexico City.
However, the release did note there was a “first part of the investigation” and I did find the publication of this prior study. This initial study definitely has interesting research, but there are some major flaws to the paper itself that make me question the strength of the study and the validity of its claim, though it is important to note that the reported results of this study presented in the press release and the results in the published paper itself do not match exactly.
In the initial study, 30 HPV+ women from Veracruz, Mexico were treated with photodynamic therapy (PDT) in which a gel form of a compound called 5-ALA was spread over the uterine-cervix and induced the formation of protoporphyrin IX (PpIX) which fluoresces in damaged tissues and allows the researches to target that specific area for irradiation with a 635nm laser (in the red spectrum). Irradiation causes a photochemical reaction resulting in production of reactive oxygen species (ROS), which has a toxic effect on the tumor cells; it also triggers the inflammatory response, which will increase the immune response in that area. In theory, if the cells with HPV could be destroyed, the presence of the virus could be eliminated (question remains about dormant viral DNA, but that’s beyond the scope of this article).
So the basic principle of the study is good and certainly very interesting, but it doesn’t seem that it was well-executed or well-presented.
If I had to grade the paper, I’d like to say “E for Effort,” but I would give it a C due to inconsistencies, poor communication, and failure to show their work. When it comes to science, a researcher can claim they have discovered something, but their words mean very little without supporting evidence. A really great study includes plenty of graphs, statistics, figures, and tables in order to support a claim, the goal being to leave the reader little room for doubt. This paper does have some of these items, but it fails to include those that would be most compelling.
Perhaps this newest phase of the study that include the 29 women in Mexico City included better supporting evidence, but unfortunately it wasn’t published. Because of the issues found within the pilot study, I am hesitant to believe they have indeed eliminated HPV in these patients as they said, at least until I can take a peak at the published data. So as much as we would all love for this to be true, it seems prudent to not get your hopes up too much just in case.
Below is the compiled list of my concerns regarding the original paper:
- Failure to Define Study Goals Explicitly
The pilot study entitled “Effectiveness of Photodynamic Therapy in Elimination of HPV-16 and HPV-18 Associated with CIN I in Mexican Women,” so one would assume the paper is about using this photodynamic therapy (PDT) to eliminate the virus. However, the FIRST line of the paper states, “This study aimed to determine the effectiveness of [PDT]… in elimination of premalignant cervical lesions in Mexican patients with [HPV] infection and/or cervical intraepithelial neoplasia (CIN).”
So is the goal to eliminate HPV or the resulting lesion? Either is a great goal, but a study needs to be consistent.
- Failure to define study groups consistently
After they chose the 30 women from the 448 gynecological samples, they say they divided them into two groups based on diagnostic results:
1) Those with CIN who had high-risk HPV (CIN + HPV)
2) Those who had HPV, but lacked evidence of HPV-induced damage (HPV+)
Then later in the results they measure the effectiveness amongst four groups: HPV+, CIN, CIN+HPV, and those with Vaginosis.
Let’s not think about the group with bacterial vaginosis at the moment. Generally one would think that “CIN” by itself means a patient who has CIN, but doesn’t have the virus.
Where did these patients come from? According to Table 2 ALL of their patients have HPV, either HPV16 or HPV18. So it’s initially unclear exactly to who is included in the “CIN” group.
In the results, they reference the “42% of the 12 patients with CIN I showed complete response with lesion regression,” (CIN I refers to the level of cervical intraepithelial neoplasia observed in the histological diagnostic). Only then does it one realize they have actually split the focus of this CIN + HPV group into showing the results of lesion regression (focusing on the CIN) and showing the HPV elimination in this group (focusing on the HPV). It’s an understandable choice but poorly executed.
The problem comes down to a failure of variable choice in the graph. They decided to use a binary scale to measure the effectiveness, which I think was a poor choice. It would have been clearer had they used multiple graphs perhaps.
- What is HPV-induced damage?
Based on the study, I would have initially suspected that those women in the second group (HPV+) would indicate they don’t have lesions. However, in order to use this treatment, they used the diameter of a target lesion to determine the total area to be irradiated and the total treatment time. So do all patients show lesions? If so, would those lesions not count as HPV-induced damage? If they don’t, it needs to be expressed what are the parameters of these lesions they are using that do and don’t count as damage. By the way, according to Table 1, yes, all patients have lesions.
- Use of binary results is limiting, and it’s the only Results graph (not including patient symptoms)
The reason it is a poor choice is often scientific results are on a scale. So a simple Yes or No is a little vague and doesn’t indicate how many patients showed even a little bit of improvement in CIN remission.
It also leaves no room for understanding which of the patients responded better to treatment. There were patients with lesions as small as 1.6 cm, to as large as 3.91 cm. Which patients showed lesion clearance, just the small lesions or big lesions, too? Did the size influence HPV elimination? Is there a correlation between the severity of CIN I and the lesion improvement or even how long it took for lesions to clear? Was there an association between HPV types and clearance?
- Failure to show proof of viral elimination
In the results section they say they eliminated the HPV infected as proved by PCR, but they didn’t include that data at all. PCR, or polymerase chain reaction, is a test in which a sample of DNA is tested to show if they have a specific gene present in the sample. In this case they are testing for the DNA of the HPV Virus.
It would be more convincing if they had included a figure showing some of patients’ samples of the PCRs initially showing them HPV+ and then showing them HPV- post-PDT treatment.
- Failure to connect patients with results
They do show the differences in irradiated areas, but don’t connect them with the patients’ diagnostics. Which patients had HPV alone and which had evidence of CIN I? Which patients showed improvement? Which patients had HPV16 and which had HPV18? Unfortunately there were only two patients HPV16 which is insignificant, but it could be of interest, even if statistically insignificant.
NOTE: They shouldn’t have claimed elimination of HPV16 in the title of the article if they didn’t prove those two patients became HPV negative after treatment.
- A couple of clerical errors
In a couple of places they noted the laser wavelength was 635nm, then in Figure 1 they write 630nm. Which is it? I’m assuming 630nm was just a typo, but reproducibility is key in scientific studies, so if the laser was ever changed, it needs to be noted. Also in the formula for Total Light Dose, they write m2 instead of cm2; nit-picky, I know, but it’s worth pointing out in case they ever want to correct it.
- Statistical mistake in Discussion
They note there were no cases of CIN2/3 because of the age range of 25-40 they used, because “according to the natural history of [HPV], high-grade lesions are more likely to be found in women over 40.” Other studies specify the median age for CIN2/3 is 34 years old, which falls directly in their chosen age range. Granted this particular study was looking at Norwegian women so it could be different, but I would be surprised if the average more than 6 years older. So their statement may be incorrect. More likely the lack of CIN2/3 patients was due to small sampling size; this study only included 30 women.
- Small sample size
All studies with small numbers should be taken with a grain of salt simply because enough replications of their results have not occurred. After they perform these repetitions and get the same results in many, many more women, their claim that they have indeed found a method to eliminate or fight the viral infection will be much more credible.
Alvarado, E. M., Peralta, M. O., Vázquez, A. M., Guzmán, L. A., Petrone, M. E., Mar, Z. I., . . . Gallegos, E. R. (2017). Effectiveness of Photodynamic Therapy in Elimination of HPV-16 and HPV-18 Associated with CIN I in Mexican Women. Photochemistry and Photobiology,93(5), 1269-1275. doi:10.1111/php.12769
Nygård, J. F., Nygård, M., Skare, G. B., & Thoresen, S. Ø. (2005). Screening histories of women with CIN 2/3 compared with women diagnosed with invasive cervical cancer: A retrospective analysis of the Norwegian Coordinated Cervical Cancer Screening Program. Cancer Causes & Control,16(4), 463-474. doi:10.1007/s10552-004-6295-z