Increased human papillomavirus viral load is correlated to higher severity of cervical disease and poorer clinical outcome: A systematic review

Recently, Seth-Frerich Fobian and colleagues published a systematic review titled "Increased human papillomavirus viral load is correlated to higher severity of cervical disease and poorer clinical outcome: A systematic review" in the Journal of Medical Virology. The findings suggest that HPV viral load is correlated to clinical outcome, and may become an important biomarker for treatment selection and response monitoring for cervical cancer.

Measurement of viral load is not a standard procedure, nor is there consensus on the methodology or quantitative values. Despite these differences, there is a clear convergence of results from both cytology and tissue samples collected for screening/referral or diagnostic purposes, that a viral load is an independent prognostic factor, next to HPV genotype, HPV co-infection, genome integration status, and other clinical factors. Future studies on viral load need more standardization and alignment of techniques and result reporting, to improve comparability and for practical use as a prognostic and predictive marker of cervical cancer and its precursors.

RESULTS

A search was conducted in the PubMed, EMBASE, and Web of Science databases using medical subject headings such as "Cervical cancer," "Cervical Intraepithelial Neoplasia," "Human Papillomavirus," and "Viral Load" in titles, abstracts, or keywords. A total of 2,331 articles were retrieved, and after screening, 85 articles were eligible for data extraction.

PRISMA flow chart. A systematic review of the viral load of HPV in cervical lesions and cancers, and its correlation to clinical outcome. HPV, human papillomavirus; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyzes.

• Detection and quantitation of HPV viral load

There is a large variation in methodology to measure HPV viral load. Common techniques used are (quantitative) polymerase chain reaction ((q)PCR), hybrid capture 2 (HC2), and in situ hybridization (ISH). Moreover, there is also little consensus on how to quantify and report viral load, neither within nor between the different techniques.

• Higher viral load correlates to more severe disease and worse clinical outcome

The majority of studies (72/85 = 84.7%) used cytology samples collected for screening/referral purposes, from 172 227 women enrolled. The remaining studies (13/85 = 15.3%) used tissue samples taken for diagnostic purposes, from 1519 women.

In general, there is substantial disparity in the sampling purposes, testing methods, and viral load quantification (“high,” “medium,” and “low”) as a prognostic marker in cervical cancer. However, despite these differences, the vast majority (87.1%; 74/85) of studies found a trend towards poor prognoses with higher viral loads, which includes CIN grades, or higher stages (disease severity) of cancer, worse overall survival, disease-free survival, progression-free survival, or higher recurrence rates.

Illustration of data collection and the correlation between the viral load of HPV and clinical outcomes. HPV, human papillomavirus; HR, hazard ratio, n.s., nonsignificant; OR, odds ratio.

• Cytology samples: Higher viral load correlates to disease severity

Of the 72 articles listed which studied cytology samples taken for screening/referral purposes, 62 (86.1%) noted that a higher viral load was linked to (a higher) CIN grade or cervical cancer stage.

In two of these studies, no correlation between viral load and CIN grade or cancer was found, and one reported the opposite correlation, noting an increase in presence of lesions among that population with decreased presence of HPV16. This is, in fact, particularly unusual given that 27/72 (37.5%) articles specifically cite that especially HPV16 is primary driver of the observed increase in severity.

In 12.5% (9/72) of the above articles, cytology samples were used to investigate a potential link between viral load and clinical outcomes, including disease-free survival and recurrence rate. A trend towards a less severe clinical outcome was linked to a higher viral load in 7 out of these 9 studies, a result strikingly opposite to the majority. A possible explanation for such a discrepancy could be that this sampling method (cytology) should not be used to evaluate clinical outcomes such as survival of cancer patients. Moreover, it was evident that these studies, all of which measuring DNA via HC2, had categorized their “low” HPV viral load group as <100 (or even 10-fold higher), whereas many other studies have a cut off at <10. This once again emphasizes the need for consistency between studies.

- Diagnostic tissue samples: Higher viral load correlates to worse clinical outcome

In 13 studies, with a total of 1519 women enrolled, biopsies were collected from patients for diagnostic purposes to determine the cancer stage and type to decide what treatment plan they should receive. Most of these studies used qPCR-based methods to quantify and demonstrate the detected range for the viral load, and actively excluded non-cancer cells in their analyzes. In the majority (11/13 = 84.6%) of included articles, an increased viral load correlated to higher stages of cancer (median: 128 patients/study, range: 13–349 patients), and to a worse clinical outcome.

Only in one study (Siriaunkgul et al.134) was a lower viral load correlated to a worse treatment outcome (shorter disease free survival, p = 0.028). However, the cohort in this study only consisted of 21 neuroendocrine cervical cancer patients. This is a rare and aggressive type of cervical cancer. All patients were HPV18 positive, and only a few patients had co-infections with HPV16, which is notable because it is more likely for the positive correlation between a high viral load and worse clinical outcome to be observed in cases with HPV16 infections,33, 126, 134, 140 which could explain the opposite correlation observed in this study.