Research Article
Human papillomavirus confers radiosensitivity in cancer cervix: a hypothesis toward a possible restoration of apoptotic pathways based on clinical outcomes
of
Niloy Ranjan Datta*,1,2, Shalini Singh1, Piyush Kumar1,3 & Dinesh Gupta4
rP
ro
Abstract Aim: To evaluate if high baseline local human papillomavirus (HPV) titer confers radiosensitivity in cancer cervix. A hypothesis is proposed to explain the clinical outcomes. Materials & methods: 121 serial HPV titers from cervical smears of 21 patients were estimated during radiotherapy (RT) and correlated with RT dose–response curves, local response and local disease-free survival (LDFS). Results: Local response (p = 0.04) and LDFS (p = 0.06) were better in high HPV than low HPV baseline group. On multivariate analysis, RT doses for 50% tumor regression and baseline HPV titer were the only predictors for LDFS. Conclusion: Serial reductions of HPV titers following RT could restore the HPV induced temporarily downregulated p53 and pRb apoptotic pathways resulting in radiosensitivity of these tumors.
Keywords
• apoptosis • cervical cancer • HPV • p53 • pRb • radiation dose–response • radiation therapy • radiosensitivity
A
ut
ho
Human papillomavirus (HPV) has been considered as an etiological agent in cancer cervix. Although more than 95% of the cervical cancers could contain high-risk HPV genomes, it may not imply that all these cancers are caused by HPV alone [1] . Presence or absence of HPV in cancer cervix has been shown to be a key prognosticator with favorable outcomes in HPV-positive cervix tumors [2–5] . Similar observations have also been documented in head and neck cancers [6–8] . Several in vitro studies have reported that HPV-positive cells exhibit a greater degree of radiosensitivity in contrast to HPV-negative cell lines [9,10] . The underlying molecular mechanisms is a subject of active research [7,10–11] . As external beam radiotherapy (EBRT) is a primary treatment modality in cancer cervix, if HPV infection could alter the radio responsiveness of these tumors, a different therapeutic approach for this subgroup of patients could be explored. We had earlier reported that patients with cancer cervix with a high HPV (HHPV) titer in their cervical smears achieved a significantly better local control and survival outcomes [5] . In the present study, we evaluated the serial clinical tumor regression with respect to their corresponding HPV titers estimated at weekly intervals during the course of EBRT. Patients were grouped into two groups as per their baseline HPV titers (HHPV and low HPV [LHPV] titers with a cut-off point of 1000 RLU/cutoff). Using radiation dose–response (D–R) parameters and the sequential changes in the HPV titers during EBRT, we examined the consecutive changes in the HPV titers during EBRT. Furthermore, a hypothesis is proposed to explain the likely molecular pathway that could result in a radiosensitivity of the tumor cells in patients having local cervical HHPV titers at the time of presentation.
Department of Radiotherapy, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India Presently at, Centre for Radiation Oncology, KSA-KSB, Cantonal Hospital of Aarau, Switzerland 3 Shri Ram Murti Smarak Institute of Medical Sciences, Bareilly, India 4 National Clinical Reference Laboratory, Delhi, India *Author for correspondence: Tel.: +41 62 838 9559; Fax: +41 62 838 5223;
[email protected] 1 2
10.2217/FON.15.53 © 2015 Future Medicine Ltd
Future Oncol. (2015) 11(9), 00–00
part of
ISSN 1479-6694
1
Research Article Datta, Singh, Kumar & Gupta Materials & methods
and subsequently at weekly intervals during the course of EBRT.
●●Patient population
●●Assessment of the local tumor regression
& dose–response curve
Primary tumor responses during EBRT were quantified as percentage regression with respect to their pretreatment size at weekly intervals during EBRT. D–R curves were plotted for each of the 21 patients. Scatter plots with EBRT dose and corresponding tumor regressions were fitted. Tumor control doses (TCD; in Gy) for 50% (TCD50) and 80% (TCD80) regressions were estimated for each patient from the individual D–R curves [14] . Patients who were completely free of any locoregional disease as evident on clinical and radiological examinations and Papanicolaou smear tests were considered to be locally free of disease and were censored for the computation of local disease-free survival (LDFS).
●●HPV titer estimation
ro
of
In total, 21 patients of squamous cell cancer cervix were enrolled for the pilot study. Pretreatment work up included clinical, hematological, biochemical and radiological investigations. Patients were staged as per FIGO guidelines. They received 50 Gy of EBRT in 5 weeks at 2 Gy/fraction followed by three applications of high-dose-rate intracavitary brachytherapy of 6 Gy each at weekly intervals prescribed at point A. None of the patients were subjected to surgery or chemotherapy. Since the treatment offered was a part of the routine treatment as per the departmental protocol, no additional approval from ethical committee was required. However, all patients were informed of the study and consent taken for weekly collection of cervical smears for HPV evaluation.
A
ut
ho
rP
The methodology of HPV titer estimations have been described in details in our earlier publication [5] . In brief, serial cervical smears were taken from the cervical tumor before EBRT and also at weekly intervals during the course of EBRT. HPV DNA was quantified using the HPV detection kit – DNA Hybrid Capture II kit (Digene, USA; presently, Qiagen, USA). It is a chemiluminescent assay based on sandwich hybridization of the denatured viral DNA with specific RNA probes containing 13 oncogenic HPV types (types 16, 18, 31, 35,39, 45, 51, 52,56, 58,59 and 68). It is therefore not possible technically to differentiate the different HPV types using this test. Chemiluminescence emitted was measured in relative light units (RLU) with the intensity of light emitted being proportional to the amount of target (HPV) DNA in the sample [12] . Test samples were classified as positive, if the ratio of the sample RLU and the positive control were equal to or greater than 1. The RLU/cutoff was a relative estimate of the HPV viral titers, compared with the positive control and according to the manufacturer’s protocol equals to 1.0 pg/ml [13] . The RLU/cutoff estimates in this study could be considered as a ‘HPV relative titer index,’ representing the ratio between RLU of the patient’s sample and the positive control. Those having a baseline HPV titer of more than 1000 RLU/cutoff were classified into HHPV titer group while those up to 1000 RLU/cut off were grouped as LHPV titer groups. In all patients, HPV titers were estimated prior to the start of the EBRT
2
Future Oncol. (2015) 11(9)
●●Statistical analysis
Independent student’s two-tailed ‘t-test’ tested the differences in means while chi-square test was used for categorical variables. Individual and subsequently grouped (for HHPV and LHPV) D–R curves were fitted to compute the TCD50 and TCD80 values. The mean difference between HPV titers for HHPV and LHPV at each week of EBRT were compared using independent student’s t-test. LDFS was computed from the date of registration and Kaplan–Meir estimates were analyzed using log rank test. Patients who continued to be locally disease free were considered censored for the corresponding survival end point. All survival analyses were based on ‘worst case scenario.’ Multivariate analysis using Cox’s regression was computed for the LDFS using the demographic variables and TCDs using backward stepwise regression analysis. Results ●●Baseline HPV titers & outcomes
A total of 121 HPV samples and tumor regressions estimates were processed from the 21 patients. Five samples were excluded due to inadvertent processing problems. All the patients had squamous cell cancers. A total of 12 patients belonged to HHPV while nine in LHPV. There were no differences in the age, stage distribution, EBRT doses and overall EBRT durations between the two groups (Table 1) [5] . Baseline HPV for the entire 21 patients ranged from 0.81 to 3966.1 RLU/cutoff (median:
future science group
Human papillomavirus confers radiosensitivity in cancer cervix
Research Article
Table 1. Demographic parameters of patients belonging to low human papillomavirus baseline and high human papillomavirus baseline groups. All patients (n = 21)
LHPV group† (n = 9) HHPV group† (n = 12) p-value‡
Age (mean ± standard deviation), years FIGO stage: –I – II – III – IV Base line HPV titer (mean ± standard deviation), RLU/cutoff Percentage change in HPV titer at end of teletherapy (mean ± standard deviation) Teletherapy dose (mean ± standard deviation), Gy Overall treatment time of teletherapy (mean ± standard deviation), days
52.3 ± 8.5 2 12 6 1 1264.4 ± 1148.2 -79.6 ± 29.4
51.0 ± 7.9 1 4 4 0 300.1 ± 336.2 -83.5 ± 23.5
53.4 ± 9.2 1 8 2 1 1987.6 ± 991.8 -76.6 ± 33.8
0.535§ 0.453¶ 1000. p-values for LHPV vs HHPV group. § t-test. ¶ Chi-square. ¶ HHPV: High human papillomavirus; LHPV: Low human papillomavirus.
of
Parameters
†
●●External radiotherapy dose–response
curves & parameters for LHPV & HHPV groups
The EBRT D–R curves were plotted for each individual patient (r2 = 0.84 for all patients, r2 = 0.78 for LHPV and r2 = 0.90 for HHPV) (Figure 2) . For the entire cohort of 21 patients, the TCD50 ranged from 24.7 Gy to 54.5 Gy (mean ± SD: 36.1 ± 7.4 Gy) while TCD80 varied from 36.2 Gy to 68 Gy (mean ± SD: 47.7 ± 6.9 Gy). Mean TCD50 for LHPV and HHPV groups were 38.5 Gy and 34.3 Gy, respectively, while the corresponding TCD80 for these groups were 50.6 Gy and 45.5 Gy, respectively. Thus, mean TCD50 and TCD80 were nearly 10% higher for the LHPV compared with HHPV patients. Using the median values of 35.4 Gy and 46.1 Gy as estimated for TCD50 and TCD80 respectively, patients were divided into two groups for each TCDs (TCD50 : ≤35.4 Gy vs >35.4 Gy; TCD80 : ≤46.1 Gy vs >46.1 Gy) to explore the effect of TCD on LDFS.
A
ut
ho
rP
1129.9 RLU/cut off). For LHPV, the baseline HPV varied from 0.81 to 815.8 RLU/cut off (median: 135.7 RLU/cut off) while in HHPV, values ranged from 1125.9 to 3966.1 RLU/cut off (median: 1649.4 RLU/cut off) (p35.4 Gy) and TCD80 group (≤ vs >46.1 Gy). Of these, the variables that were significant or showed a trend toward significance on univariate analysis for
www.futuremedicine.com
3
Research Article Datta, Singh, Kumar & Gupta
2500 HHPV LHPV
1500
1000
of
HPV titers (RLU/cut-off)
2000
0.00
1.00
2.00 3.00 Week from start of RT
rP
0
ro
500
4.00
5.00
ho
Figure 1. Changes in the human papillomavirus titers during the 5 weeks of external radiotherapy in high human papillomavirus and low human papillomavirus group. Error bars represent ±1 SE. HHPV: High human papillomavirus; LHPV: Low human papillomavirus.
A
ut
LDFS included – stage group (median LDFS for I + II vs III + IV as 6.4 vs 1.6 years, p = 0.047), baseline HPV (median LDFS, LHPV vs HHPV as 1.6 vs 6.4 years; p = 0.065) (Figure 3) and TCD50 (median LDFS not reached vs 1.6 years for TCD50 ≤ vs >35.4 Gy; p = 0.01). Multivariate Cox’s proportional hazard model was performed to identify the possible predictors for LDFS. Only, TCD50 and baseline HPV were retained in the model as significant predictors of LDFS. A lower TCD50 (exp(β) 0.222, p = 0.016; 95% CI: 0.065–0.756) and a higher baseline HPV titer (exp(β) 3.070, p = 0.065; 95% CI: 0.933–10.018) predicted for a superior LDFS. Discussion Neoplastic transformations induced by HPV have been attributed to the E6 and E7 oncoproteins, which interfere with the functioning of tumor suppressor genes – p53 and retinoblastoma (pRb) genes, respectively [1,10,15–17] . These genes, which normally control the cell signaling pathway to regulate the cell cycle, monitor
4
Future Oncol. (2015) 11(9)
and protect genomic integrity are inhibited by E6 and E7, leading to a loss of intrinsic control mechanism. Thus, based on the plausible causative theory, it would be imperative to assume that HPV, the key factor in the etiopathogenesis of cancer cervix should also attribute to greater virulence to the disease. However, clinical reports are mostly contrary to this [2–6] . ●●Modulation of radiosensitivity by E6 & E7
of HPV
A number of in vitro studies have investigated the possible role of E6 and E7 of HPV in modulating the radiosensitivity. The results of these studies were variable and have been well summarized by Vozenin et al. [10] . The possible pathways that E6 and E7 could modulate radiosensitivity are through p53, p73, p21PI3k/ AKT or MAPKs pathways. Various squamous cell carcinoma cell lines have been also shown to be radiosensitized by HPV 16 oncoprotein E6*1 [9] . Wang et al. reported that radiation cell kill in cervix cancer xenografts get potentiated
future science group
Human papillomavirus confers radiosensitivity in cancer cervix
Research Article
Table 2. Changes in the human papillomavirus titers in high human papillomavirus and low human papillomavirus groups during the course of external beam radiotherapy. EBRT week
LHPV (mean ± standard deviation)
HHPV (mean ± standard deviation)
p-value (t-test)
0 1 2 3 4 5
300.14 ± 336.17 88.38 ± 129.89 659.38 ± 1209.61 537.15 ± 941.70 216.61 ± 346.95 61.90 ± 152.65
1987.57 ± 991.78 1946.07 ± 1169.05 1761.92 ± 1416.04 1644.93 ± 1461.08 1051.19 ± 1235.25 527.40 ± 785.82
1000 RLU/cutoff) compared with those with a lower HPV titer (≤1000 RLU/cutoff). ●●
Local response and local disease-free survivals were better in patients with HHPV.
●●
The slope of HHPV dose–response curve was steeper, resulting in 10% reduction in mean radiotherapy doses (tumor
●●
Multivariate analysis for local disease-free survival retained only TCD50 and baseline HPV as outcome predictors.
Discussion
Higher HPV baseline titer in cancer cervix results in better outcomes to radiotherapy, indicating a clinically
A
●●
ut
control doses [TCD]) for 50 and 80% tumor regressions.
radiosensitive cervical tumor subpopulation. ●●
HHPV could temporarily downregulate the p53 and pRb pathways resulting in neoplastic transformation in the cervix. During external radiotherapy, the reductions in HPV could result in repression of E6 and E7 oncoproteins, leading to restoration of the dormant p53 and pRb apoptotic pathways. This could lead result in a radiosensitive subpopulation.
●●
By contrast, carcinogenesis in low HPV could be due to multigenic complex mutations rendering these tumors more aggressive. Radiotherapy may not influence the resumption of apoptosis in these, resulting in a relatively radioresistant tumor subpopulation.
References Papers of special note have been highlighted as: • of interest; •• of considerable interest 1
8
Zur Hausen H. Papillomaviruses in the causation of human cancers – a brief historical account. Virology 384(2), 260–265 (2009).
•• A very comprehensive minireview on the role of papillomavirus as a causative factor for different human cancers including cancer cervix. 2
Rodríguez-Carunchio L, Soveral I, Steenbergen R et al. HPV-negative carcinoma
Future Oncol. (2015) 11(9)
of the uterine cervix: a distinct type of cervical cancer with poor prognosis. BJOG 122(1), 119–127 (2015). •
Human papillomavirus (HPV) negative cancer cervix has been shown to be of worse prognosis with increased risk of
future science group
Human papillomavirus confers radiosensitivity in cancer cervix disease progression and mortality.
•
6
•
7
Authors previous paper on which the present study is based showing that patients with high HPV titers have better survival outcomes. Friedman JM, Stavas MJ, Cmelak AJ. Clinical and scientific impact of human papillomavirus on head and neck cancer. World J. Clin. Oncol. 5(4), 781–791 (2014). A comprehensive review of the current literature on the clinical outcomes with HPV in head and neck cancers along with the likely molecular differences between the HPV-positive and -negative patients.
9
Unraveling the biology of human papillomavirus (HPV) related tumours to enhance their radiosensitivity. Cancer Treat. Rev. 36(8), 629–636 (2010).
HNSCC cell lines positive for HPV and p16 possess higher cellular radiosensitivity due to an impaired DSB repair capacity. Radiother. Oncol. 107(2), 242–246 (2013). 20 Kimple RJ, Smith MA, Blitzer GC et al.
Enhanced radiosensitivity in HPV-positive head and neck cancer. Cancer Res. 73(15), 4791–4800 (2013). 21 Psyrri A, Cohen E. Oropharyngeal cancer:
of
have we learned from human papillomavirus - positive tumours? Trying to connect data about biomarkers among human papillomavirus-related squamous cell carcinomas. J. Clin. Oncol. 28(20), e340–e341 (2010).
clinical implications of the HPV connection. Ann. Oncol. 22(5), 997–999 (2011).
12 Lorincz AT. Hybrid capture method for
detection of human papillomavirus DNA in clinical specimens; a tool for clinical management of equivocal Pap smears and population screening. J. Obstet. Gynaecol. Res. 22(6), 629–636 (1996). •
Gives details of the DNA hybrid capture method for detection of HPV as used in this study.
22 Tan S, Hougardy BM, Meersma GJ et al.
et al. Accuracy of human papillomavirus testing in primary screening of cervical neoplasia: results from a multicenter study in India. Int. J. Cancer 112(2), 341–347 (2004).
•
of tumour suppressor p53 by differentially regulating pro and anti-p53 pathways in HPV-18 infected cervical cancer cells. Oncogene 31(2) 173–186 (2012). 24 Huang H, Huang SY, Chen TT, Chen JC,
Chiou CL, Huang TM. Cisplatin restores p53 function and enhances the radiosensitivity in HPV16 E6 containing SiHa cells. J. Cell Biochem. 91(4), 756–765 (2004).
Application of the DNA hybrid capture method for screening of HPV in a similar population of patients.
Lohaus F, Linge A, Tinhofer I et al. HPV16 DNA status is a strong prognosticator of loco-regional control after postoperative radiochemotherapy of locally advanced oropharyngeal carcinoma: results from a multicenter explorative study of the German Cancer Consortium Radiation Oncology Group (DKTK-ROG). Radiother. Oncol. 113(3), 317–323 (2014).
position and steepness of radiation doseresponse curves. Int. J. Radiat. Biol. 71(5), 531–542 (1997). Gives details of the parameters to quantify the radiation dose–response relationship.
15 Johannsen E, Lambert PF. Epigenetics of
human papillomaviruses. Virology 445(1–2), 205–212 (2013). 16 Stanley M. Pathology and epidemiology of
HPV infection in females. Gynecol. Oncol. 117(2 Suppl.), S5–S10 (2010). 17 Nguyen HP, Ramirez-Fort MK, Rady PL.
The biology of human papillomaviruses. Curr. Probl. Dermatol. 45, 19–32 (2014).
Human papillomavirus 16 E6 RNA interference enhances cisplatin and death receptor-mediated apoptosis in human cervical carcinoma cells. Mol. Pharmacol. 81(5), 701–709 (2012).
23 Saha B, Adhikary A, Ray P et al. Restoration
13 Sankaranarayanan R, Chatterji R, Shastri SS
•
future science group
19 Rieckmann T, Tribius S, Grob TJ et al.
11 Vozenin MC, Bourhis J, Deutsch E. What
14 Bentzen SM, Tucker SL. Quantifying the
Pang E, Delic NC, Hong A, Zhang M, Rose BR, Lyons JG. Radiosensitization of oropharyngeal squamous cell carcinoma cells
of tumour radiotherapy by a radiationinducible oncolytic and oncoapoptotic adenovirus in cancer cervix xenografts. Int. J. Cancer 130(2), 443–453 (2012).
•• Summarizes the current understanding of radiosensitivity and HPV-related tumor biology.
Lassen P. The role of human papillomavirus in head and neck cancer and the impact on radiotherapy outcome. Radiother. Oncol. 95(3), 371–380 (2010).
A
8
Datta NR, Kumar P, Singh S, Gupta D, Srivastava A, Dhole TN. Does pretreatment human papillomavirus (HPV) titers predict radiation response and survival outcomes in cancer cervix? – A pilot study. Gynecol. Oncol. 103(1), 100–105 (2006).
18 Wang H, Song X, Zhang H et al. Potentiation
ro
5
Wang C-C, Lai C-H, Huang Y-T, Chao A, Chou H-H, Hong J-H. HPV genotypes predict survival benefits from concurrent chemotherapy and radiation therapy in advanced squamous cell carcinoma of the cervix. Int. J. Radiat. Oncol. Biol. Phys. 84(4), e499–e506 (2012).
•• A comprehensive review on the biology of the HPVs.
10 Vozenin MC, Lord HK, Hartl D, Deutsch E.
rP
4
One of the key initial papers that showed that low HPV viral load was associated with a worse prognosis in cancer cervix.
by human papillomavirus 16 oncoprotein E6*1. Int. J. Radiat. Oncol. Biol. Phys. 79(3), 860–865 (2011).
ho
•
Kim JY, Park S, Nam BH et al. Low initial human papilloma viral load implicates worse prognosis in patients with uterine cervical cancer treated with radiotherapy. J. Clin. Oncol. 27(30), 5088–5093 (2009).
ut
3
Research Article
•
Restoration of p53 function by cisplatin in HPV-positive cell lines.
25 Kranjec C, Massimi P, Banks L. Restoration
of MAGI-1 expression in human papillomavirus positive tumor cells induces cell growth arrest and apoptosis. J. Virol. 88(13), 7155–7169 (2014). 26 Bennett C, Kuhn AE, Haverkos HW. Human
papillomavirus and tar hypothesis for squamous cell cervical cancer. J. Biosci. 35(3), 331–337 (2010). 27 National Cancer Institute. NCI issues clinical
announcement on cervical cancer: chemotherapy plus radiation improves survival. www.cancer.gov
www.futuremedicine.com
9
