Study ID

Method

Patient characteristics

Intervention(s)/ Exposure

Results

Comments

• Reference (first author and year)

• Design (Observational study)
• Conflicts of interest reported
• Setting (number of centres and country)
• Sample size 
• Duration (time of follow-up)
• Study conducted in/recruitment (year – year)
• Protocol existence reported? (clinicaltrials.gov / WHO register)

• Eligibility criteria (include age, sex, and stage of cancer)
• Baseline patient characteristics (include age, sex, and stage of cancer)

• Intervention(s)/ Exposure (and number of patients).
• Comparator(s)/ non exposure (and number of patients).

• Per outcome

Hwang, 2015

Design: Retrospective cohort study

Conflicts of interest: Bruno P. Granwehr: Research Funding: Gilead Sciences (Inst). Harrys A. Torres: Consulting or Advisory Role: Gilead Sciences, Merck, Janssen, and Research Funding: Merck, Vertex, The University of Texas MD Anderson Cancer Center. Michael J. Fisch: Employment: AIM Specialty Health and Stock or Other Ownership: Anthem and Consulting or Advisory Role: WellPoint. All other authors: No relationship to disclose.

Setting: A large US Comprehensive Cancer Center

Sample size: 245

Duration: 7 years and 4 months

Study period: January 1, 2004, through April 30, 2011

Protocol reported: Not reported

Eligibility: Adults with cervical cancer who received systemic cancer therapy (not oral chemotherapy) registered at The University of Texas MD Anderson Cancer Center. Patients included in clinical trials were excluded.

Baseline patient characteristics: Baseline characteristics were reported for the study population as a whole, which included patients with several types of cancer, and not for patients with cervical cancer separately. The overall mean age was 55 years (SD 13.5).

Intervention: HIV tested (and result): We determined the test dates and results for HIV testing, defined as ordering of an HIV-1/2 antibody test by enzyme linked immunosorbent assay and/or confirmatory Western blot testing after registration at MD Anderson. We defined the period of initiation of cancer therapy as the period from 2 months before the first systemic cancer therapy administration until receipt of the second systemic cancer therapy administration.

Comparator: Not HIV tested

Of 245 patients with cervical cancer, 23 were tested for HIV. None were found to be HIV positive.

Discussion: the NCCN cervical cancer guideline panel had major disagreements about HIV testing, and thus, routine screening is not supported by the NCCN. The lack of guidance from national oncology societies may have influenced the low rate of HIV testing among patients with cervical cancer in our study (9%), which was similar to that in another study of patients with cervical cancer (11%).

Because not all positive HIV test results were confirmed by Western blot testing, the HIV test results in this study need to be interpreted with caution. Patients could have had HIV testing before our study testing period or outside of MD Anderson, and providers may have elected not to repeat HIV testing before initiating cancer therapy, even among patients with known HIV infection.

Bogers, 2021 (1)

Design: Retrospective cohort study

Conflicts of interest:  The authors have declared that no competing interests exist.

Setting: Five hospitals in the region of Amsterdam, the Netherlands

Sample size: 1,075

Duration: 5.5 years

Study period: 01/2015-06/2020

Protocol reported: Yes, available online.

Eligibility: Patients ≥18 years not known to have HIV, diagnosed with  CIN-3/ICC during 2015-2020 at one of the study sites.

Baseline patient characteristics: Median age 40 years (IQR 32-51).

Intervention:  HIV tested (and result):  tested for HIV ≤3 months before or after  CIN-3/ICC diagnosis

Comparator:  Not HIV tested

Of 1,075 patients with  CIN-3/ICC, 46 (1.3%) were tested for HIV.  None were found to be HIV positive.

In initial search: Conference abstract, .

Full paper and  detailed data  were provided by the author. (1)

Bogers, 2021 (2)

Design: Systematic review and meta-analysis

Conflicts of interest:  Dr. Bogers reports grants from Aidsfonds, grants from H-TEAM. Dr. Schim van der Loeff reports other from Merck, non-financial support from Stichting Pathologie Onderzoek en Ontwikkeling (SPOO). Dr. de Bree reports grants from AIDS Fonds, grants

from Stichting Amsterdam Dinner Foundation, grants from Gilead Sciences, grants from Janssen Pharmaceutica, grants from ViiV Healthcare, grants from ZonMW, grants from M.A.C AIDS Fund, grants and other from Gilead Sciences.

Dr. Reiss reports grants from aidsfonds, grants

from Stichting Amsterdam Dinner Foundation, grants from Gilead Sciences, grants from Janssen Pharmaceutica, grants from ViiV Healthcare, grants from ZonMW, grants from M.A.C AIDS Fund, grants and other from Gilead Sciences, grants

and other from ViiV Healthcare, grants and other from Merck.

Dr. van Bergen reports grants from RIVM and reports being a member of the board of the H-TEAM. Dr. Geerlings reports grants from Aidsfonds.

Serach date and data bases:

Embase and Ovid MEDLINE were searched up to November 20th, 2020

Setting: Western countries  (Western Europe, USA, Canada, Australia, New Zealand, and Japan)

Sample size:  2,878.  Median number of subjects was 489 (IQR 245-583).

Number of publications: 57

23 full text articles 34 other (abstract short communication, correspondendce) ,

Duration: n/a

Study period: The publication years of included studies ranged from 2011-2020. Study periods of included studies ranged from 2002-2017.

Protocol reported:  Yes, available online.

PROSPERO, registration number CRD42020160243

Eligibility:  Studies reporting HIV test ratios among patients of 18+ years (directly available or through calculation with presented data), all settings (e.g. primary care (PC), hospital care, registry surveillance), and all publication types (e.g. research article, abstract, correspondence)

were eligible for inclusion. No language restrictions were applied. Studies among persons known HIV positive, with unconfirmed cervical cancer (CC) or cervical intraepithelial neoplasia (CIN) grade 2+, and studies with a sample size <10 were excluded. Studies with data on HIV testing before 2009 only were excluded, as IC-guided testing was

globally implemented around 2009.

Baseline patient characteristics: not reported for patients with  cervical cancer (CC) or cervical intraepithelial neoplasia (CIN) grade 2+ separately.

Intervention: HIV tested

Comparator:  Not HIV tested

Of 2,878 patients from 10 studies with  cervical cancer (CC) or cervical intraepithelial neoplasia (CIN) grade 2+, 600 (20.9%) were tested for HIV. Of the four studies reporting on positivity, three reported 0% and one 0.2% positivity.

Large differences in HIV test ratios between studies concerning the same IC were observed. Some outliers were studies with a high risk of bias, but among studies with low risk of bias, considerable variation

was still observed. An explanation may be the difference in design of studies and how being tested for HIV was defined: some studies assessed evidence of any HIV testing, while others had a set

timeframe around IC diagnosis to assess IC-guided testing.

Raben, 2015

Design:  Retrospective cohort

Conflicts of interest: The authors have declared that no competing interests exist.

Setting:  Healthcare centres/hospitals across the four regions of Europe: North, East, South and West

Sample size: 583

Duration:  The median period of time for retrospective data collection was 1.5 years.

Study period:  Data was collected retrospectively from May 2013

Protocol reported:  Not reported

Eligibility: All consecutive patients > 18 and <65 years of age, not known to be HIV positive, who had presented within the previous year or the last 100 consecutive patients or more. Patients already known HIV positive before cervical cancer diagnosis were excluded

Baseline patient characteristics:  Baseline characteristics were reported for the study population as a whole, which included patients with several conditions, and not for patients with cervical cancer separately.

Intervention:  HIV tested (and result) - Participating centres reviewed retrospectively how many patients presenting with  cervical cancer were tested for HIV.

Comparator:  Not HIV tested

Of 583 patients with cervical cancer, 444 (76.2%) were tested for HIV. Of these, 1 (0.2%) was HIV positive

The data presented are the summary aggregate data from the audits. As individual patient-level data were not available, it has not been possible to take patient variation and medical history into account in the analysis. It is possible that clinics have over-reported HIV testing. It is also possible that clinicians have targeted the HIV testing offered to those at perceived greatest risk of testing positive, which may mean that our rate of testing positive is overestimated and that the number of missed diagnoses would be smaller than estimated.

Mosimann, 2014

Design: Retrospective cohort

Conflicts of interest: The authors have no conflicts of interest to declare.

Setting: Lausanne University hospital, Switzerland

Sample size: 57

Duration: 10.5 years

Study period: January 2002 - July 2012

Protocol reported:  Not reported

Eligibility: Patients aged ≥ 18 years treated for invasive cervical cancer

Baseline patient characteristics: Median age and IQR: Tested: 49 (36-58), Not tested: 54 (44-65)

Intervention: HIV tested (and result) - HIV testing was considered as related to oncology work-up if performed within 90 days before and 90 days after the cancer diagnosis date. Patients were classified as HIV ‘tested’ upon identification of [1] documented testing in the hospital database or [2] a reference to testing performed elsewhere.

Comparator:  Not HIV tested

Of 57 patients with invasive cervical cancer, 6 were tested for HIV. One patient (16.6%) was HIV positive.

Assuming all untested patients were HIV negative, HIV seroprevalence among ICC patients was 1.7%. The low testing rate in ICC patients is surprising, given that  ICC  is  an  ADC.  The  authors  of  the  HIV  Indicator Diseases across Europe Study (HIDES 1) recently concluded that HIV screening in patients with cervical cancer of any stage is cost effective and should  be  recommended. However, neither European  nor American ICC  guidelines currently mention testing.

Qureshi, 2017

Design: Prospective cohort study

Conflicts of interest: The authors have declared that no competing interests exist.

Setting: Birmingham Women’s Hospital colposcopy clinic, UK

Sample size: 252

Duration: almost 2 years

Study period: between September 2013 and June 2015

Protocol reported: Not reported.

Eligibility: Women with High-Grade cervical dyskaryosis who could understand English and not known to be HIV +

Baseline patient characteristics:

Age (median) 34 [severe], 30 [moderate], 28 [mild], 28 [borderline]

Race: Caucasian 91%, Asian 6%, African 3% [severe], Caucasian 72%, Asian 17%, African 11% [moderate], Caucasian 70%, Asian 25%, African 5% [mild], Caucasian 67%, Asian 25%, African 7% [moderate]

Intervention: All patients were offered HIV testing as part of their clinic management at colposcopy unit.

Comparator: a control group of women presenting with Low Grade Cervical smear abnormalities.

Of 252 patients 244 accepted to have HIV screening during their visit. None of them tested positive.

Of the 244 patients tested in the study, none were HIV positive. We believe this was secondary to HIV testing of a small proportion (7%) of women attending colposcopy services. We would have diagnosed HIV infected women, were the test offered to more patients. Birmingham Women’s Hospital antenatal clinics routinely offer HIV testing to all pregnant women. HIV infected pregnant women were diagnosed at the rate of 2.5 per 1000 (23/9080) in the calendar year of 2015–2016 (personal communication). This prevalence rate is in line with the rest of England’s

Agustí, 2016

Design: cross-sectional population based study

Conflicts of interest: The authors have declared that no competing interests exist.

Setting: 285 PC centers managed by the main healthcare provider (the Catalan Health Institute), Spain

Sample size: 615

Duration: 1.5 years

Study period: 1 January 2010 and 31 august 2012

Protocol reported: Not reported.

Eligibility: Patients between 16 and 65 years, belonged to one of the PC centres, diagnosed with an indicator condition and not known to be HIV positive.

Baseline patient characteristics:

Baseline characteristics were reported for the study population as a whole, which included patients with several conditions, and not for patients with cervical cancer separately.

Intervention: HIV test was associated with the diagnosis of an indicator condition if it was performed within 4 months of the indicator condition diagnosis date.

Comparator: Patients not tested for HIV versus patients tested for HIV, and patients tested positive versus patients tested negative

Out of 615 patient diagnoses with cervical cancer 15 were tested for HIV and 0 tested positive.

Of the 13 patients not tested for HIV at the first episode and later found to be HIV positive, 11 were men and two were women. Four patients were diagnosed with an STI at the first episode, two with herpes zoster infection, two with hepatitis B, two with seborrhoeic dermatitis, one with hepatitis C and one with cervical carcinoma.

Carlander, 2017

Design: Population-based register study

Conflicts of interest:  The authors have declared that no competing interests exist.

Setting: Counties of Stockholm and Gothenburg, Sweden

Sample size: 62874

Duration:  24 years

Study period: 1990 - 2014

Protocol reported: Not Reported

Eligibility: all women born between 1940 and 1990, residing in the counties of Stockholm or Gothenburg at some point between 1990 and 2014 with at least one cervical cytology or histology registered in the Swedish National Cervical Screening Register

Baseline patient characteristics: not reported

Intervention: None, to estimate the time of HIV acquisition, they applied a CD4+ trajectory model to estimate the time between HIV acquisition and HIV diagnosis.

Comparator: No HIV infection

62874 women had at least one registered diagnosis of CIN2+, of these 175 women had CIN2+ diagnosed after HIV diagnosis and 62653 had no HIV diagnosis

46 women had CIN2+ diagnosed before the date of HIV diagnosis, of which 8 were estimated to have CIN2+ before estimated HIV acquisition and 38 women were defined as having undiagnosed HIV at time of CIN2+ diagnosis

The proportion of undiagnosed HIV was higher among women with CIN2+ than among women without [0.06% (95% CI 0.04–0.08) versus 0.04% (95% CI 0.04–0.04); P = 0.017)] but did not reach the threshold of 0.1%

Among migrants, the proportion of undiagnosed HIV was also higher among women with CIN2+ than among women without [0.30% (95% CI 0.20–0.43) versus 0.08% (95% CI 0.07–0.10); P < 0.001] and exceeded 0.1%.

Raben, 2019

Design: Prospective cohort study

Conflicts of interest: The HIDES study was funded by the HIV in Europe initiative which has received unrestricted funding from Gilead Sciences, ViiV Healthcare, Merck, Tibotec, Pfizer, Schering-Plough, Abbott, Boehringer Ingelheim, Bristol-Myers Squibb, GlaxoSmithKline. This does not alter our adherence to PLOS ONE policies on sharing data and materials. AM has received honoraria and consultancy fee from Gilead Sciences and ViiV Healthcare. JG has received honoraria and grants from ViiV Healthcare, MSD, Gilead Sciences and Janssen and is by 1 May 2018 employed by ViiV Healthcare. JR has received honoraria and consultancy fee from ViiV Healthcare, Merck/MSD, Gilead Sciences, Abbott, Abbvie, Abivax, Janssen. YY has received speaker and consultancy fee from Abbott, BMS, Gilead Sciences, MSD, Roche, Tibotec and ViiV Healthcare. KC has received consultancy and speaker fee from Gilead. All other authors declare no conflict of interest.

Setting:  Healthcare centres/hospitals across 20 countries the four regions of Europe: North, East, South and West

Sample size: 1341

Duration: 2.5 years

Study period: Between January 2012 and June 2014, for mononucleosis-like illness until September 2015

Protocol reported: Not reported.

Eligibility: Patients aged 18-65 years, not known to be HIV+ and presenting for care with one of the 14 selected HIV indicator conditions.

Baseline patient characteristics:  Baseline characteristics were reported for the study population as a whole, which included patients with several conditions, and not for patients with cervical cancer separately.

Intervention: All patients presenting with cervical dysplasia or cervical cancer were offered an HIV test

Comparator: Patients that tested positive for HIV versus patients that tested negative for HIV

13 out of 1341 patients presenting with cervical dysplasia or cervical cancer tested positive for HIV

Study reference

(first author, year of publication)

Bias due to a non-representative or ill-defined sample of patients?¹

(unlikely/likely/unclear)

Bias due to  insufficiently long, or   incomplete follow-up, or differences in follow-up between treatment groups?²

(unlikely/likely/unclear)

Bias due to ill-defined or inadequately measured outcome ?³

(unlikely/likely/unclear)

Bias due to inadequate adjustment for all  important prognostic factors?⁴

(unlikely/likely/unclear)

Hwang, 2015

Likely

Unlikely

Unlikely

Unlikely

Bogers, 2021 (1; Cohort)

Likely

Unlikely

Unlikely

Unlikely

Bogers, 2021 (2; SR)

Likely

Unlikely

Unlikely

Unlikely

Raben, 2015

Unlikely

Unlikely

Unlikely

Unlikely

Mosimann, 2014

Likely

(only 57 patients in 10 years)

Unlikely

Unlikely

Unclear (not extensively described – article type = short communication)

Quereshi, N., 2017

Unlikely

Unlikely

Unlikely

Unlikely

Agustí, C., 2016

Likely

(small sample, only 15 out of 615 patients with cervical cancer were tested for HIV)

Unlikely

Unlikely

Unlikely

Carlander, C., 2017

Unlikely

Unlikely

Unclear

(outcome is proportion undiagnosed HIV instead of HIV positivity test rate)

Unlikely

Raben, D., 2019

Unlikely

Unlikely

Unclear

(No differentiation between cervix cancer and dysplasia, however, showing a high HIV positivity rate and expectation that cervical dysplasia will decrease the HIV positivity rate)

Unlikely

Auteur en jaartal

Redenen van exclusie

Gupta N, 2011

Geen % hiv-positieven weergegeven

Taulo F, 2008

Geen % hiv-positieven weergegeven voor de groep met cervixcarcinoom, alleen over groep met alle gynaeocologische maligniteiten beschikbaar

Wallace S, 2001

Een review met een klein relevant stukje of kosteneffectiviteit, maar niet over % hiv-positieven

Russi A, 2018

Populatie waren mensen die kwamen voor cervixscreening en niet gediagnosticeerde cervixcarcinomen

Dottino P, 1993

Studie neemt ook patiënten mee die kwamen voor cervixscreening, geen full-text beschikbaar, uitgebreidere data beschikbaar in ref #23

Temmerman M, 1998

Geen % hiv-positieven weergegeven voor groep met cervixcarcinoom

Omire A, 2020

Populatie waren vrouwen die voor cervix screening kwamen, maar 8 hadden een afwijkend uitstrijkje

Hartney T, 2020

Conference abstract, geen full-text beschikbaar, te weinig data beschikbaar

Dartell M, 2014

Geen onderscheid welke patiënten die hiv positief zijn HSIL of LSIL hebben

Joshi S, 2001

Geen onderscheid welke patiënten die hiv positief zijn HSIL of LSIL hebben

Traore B, 2015

Geen % hiv-positieven weergegeven

Goodman A, 2000

Geen full-text beschikbaar, Geen onderscheid welke patiënten die hiv positief zijn een abnormaal uitstrijkje hebben

Abu-rustum N, 2001

Letter to the editor, geen full-text beschikbaar, te weinig data beschikbaar.

Sullivan A, 2013

Geen onderscheid gemaakt in data tussen cervixcarinoom en cervixdysplasie en anuscarcinoom en anusdysplasie

Mutyaba T, 2010

Geen onderscheid welke patiënten die hiv positief zijn HSIL, LSIL of cervixcarcinoom hebben