If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author: Address: The GI Unit, Department of Medicine, The Royal Marsden Hospital, Fulham Road, London SW3 6JJ, UK. Tel.: +44 207 811 8216; fax: +44 207 811 8107.
3-Hydroxy-methylglutaryl coenzyme-A reductase inhibitors (statins) improve survival following pelvic irradiation for cancer. Large studies suggest that patients with hypertension may have reduced gastrointestinal (GI) toxicity. Animal data suggest that statins and ACE inhibitors (ACEi) may protect against normal tissue injury. Their efficacy in humans has not been reported.
Aims/methods
To evaluate the impact of statins and ACEi on normal tissue toxicity during radical pelvic radiotherapy. GI symptomatology was recorded prospectively before radiotherapy, weekly during treatment and 1 year later using the Inflammatory Bowel Disease Questionnaire – Bowel (IBDQ-B) subset. Cumulative acute toxicity (IBDQ-B AUC) and worst score were determined. Dose, brand and duration of statin and/or ACEi usage were obtained from General Practitioners.
Results
Of 308 patients recruited, 237 had evaluable acute drug and toxicity data and 164 had data at 1 year. Acutely, 38 patients (16%) were taking statins, 39 patients (16.5%) were taking ACEi and 18 patients (7.6%) were taking statin + ACEi. Mean changes in acute scores were 7.3 points (non-statin users), 7.3 (non-ACEi users) and 7.0 (non-statin + ACEi users) compared to 4.8 points (statin users), 5.0 points (ACEi users) and 4.9 points (statin + ACEi users). Statin use (p = 0.04) and combined statin + ACEi use (p = 0.008) were associated with reduced acute IBDQ-B AUC after controlling for baseline scores (ANOVA). At 1 year, users maintained higher IBDQ-B scores than non-users in all user subgroups.
Conclusion
Use of statin or statin + ACEi medication during radical pelvic radiotherapy significantly reduces acute gastrointestinal symptoms scores and also appears to provide longer-term sustained protection.
Clinical evidence suggests that the cholesterol-lowering agents, 3-hydroxy-methylglutaryl coenzyme-A reductase inhibitors, ‘statins’, improve outcomes following therapeutic irradiation for pelvic cancers. Of three studies conducted between 2009 and 2011 in 3340 prostate cancer patients of whom 23–27% of patients reported using statins, two studies
(n = 2372) showed significant improvements in PSA relapse free survival and freedom from biochemical failure respectively in statin users versus non-users whilst a third study
of whom 9% were reportedly taking statins, a significantly improved complete pathological response rate following radiotherapy was reported in statin users and in a further study of 286 bladder cancer patients
(12% reporting statin use) improved local control was reported in statin users on univariate (but not multivariate) analysis. The mechanism of action is attributed to enhanced sensitisation of tumour cells to cytocidal irradiation.
However, statins may also be protective for normal tissues. The mechanism of action is (at least partly) via blockade of the enzyme HMGCoA reductase and thus a reduced cellular pool of mevalonate derived isoprenoid intermediates required for the activation of Rho (ras homologous) GTPases and consequently the Rho/ROCK (Rho-associated kinase) signalling pathway. Impedance results in downstream inhibition of pro-fibrotic and pro-inflammatory cytokines, attenuation of mediators of vascular damage and failure to respond to cytokine or growth factor mediated stimulation. In-vitro investigations have clearly demonstrated the anti-inflammatory,
No clinical studies have yet reported on these topics.
ACE (angiotensin I-converting enzyme) inhibitors (ACEi) may also have a role in the reduction of radiation-induced toxicity. These anti-hypertensive agents block enzymatic conversion of angiotensin I to angiotensin II playing a critical role in blood pressure homoeostasis. Animal studies focussing on the protective effects of ACEi in lung tissue yielded promising results.
has addressed the efficacy of ACEi and gastrointestinal toxicity but showed no benefit. In lung patients, two clinical studies (10 years apart) have been conducted, one
showed a clear benefit in 14% of the ACEi users (n = 146) in preventing radiation-induced lung injury in multivariate analysis, whilst the earlier study did not.
Clinical factors predicting late severe urinary toxicity after postoperative radiotherapy for prostate carcinoma: a single-institute analysis of 742 patients.
conducted between 2004 and 2011 amounting to n = 4855 prostate cancer patients have included an analysis of the impact of presence of hypertension on normal tissue toxicity with three studies
reporting a protective effect. Only one of these studies – from our unit – included patients treated for non-urological cancers and this showed reduced toxicity in hypertensive patients irrespective of tumour site.
The most likely reason for the paradoxical finding of reduced toxicity in hypertensive patients comes from the use of anti-hypertensive medication. In the five studies
Clinical factors predicting late severe urinary toxicity after postoperative radiotherapy for prostate carcinoma: a single-institute analysis of 742 patients.
Clinical factors predicting late severe urinary toxicity after postoperative radiotherapy for prostate carcinoma: a single-institute analysis of 742 patients.
(n = 742) reported that the presence of hypertension was predictive for late grade 2 gastrointestinal toxicity.
We aimed to determine whether statins and ACEi reduce normal tissue toxicity during therapeutic pelvic radiotherapy. Data were used from a mixed cohort of 308 pelvic radiotherapy patients in whom gastrointestinal toxicity outcomes had been previously captured prospectively
and the impact of individual statin and ACEi use on gastrointestinal symptoms was retrospectively analysed in the acute and late setting.
2. Methods
2.1 Study design
This single-centre study was approved by the Research and Development and Ethics committees of the Royal Marsden NHS Foundation Trust.
2.2 Eligibility criteria
Patients with histologically proven gynaecological, urological or lower gastrointestinal malignancy due to receive radical (long-course) external beam radiotherapy were invited to participate in our original study.
GI symptom scoring was carried out prospectively. Time-points were baseline (start of radiotherapy), once a week during radiotherapy and 1 year later. The 10 question Inflammatory Bowel Disease Questionnaire – Bowel (IBDQ-B) subset, which is embedded within the 32-question modified Inflammatory Bowel Disease Questionnaire (IBDQ) was used. The IBDQ-B scores a maximum of 7 points per question (maximum score: 70 ‘asymptomatic’) and a minimum of 1 point per question (minimum score: 10 ‘most severe symptoms’) with scores between 10 and 70 reflecting gradations of severity.
2.5 Toxicity data treatment
Patients with start and end of radiotherapy IBDQ-B scores were included in an acute analysis and those with additional 1 year scores included in the late analysis. Duration of radiotherapy (weeks) was compared against the availability of completed questionnaires and missing IBDQ-B scores during treatment were replaced by the last available score. Cumulative acute symptom scores (as area under the curve for IBDQ-B, i.e. ‘IBDQ-B AUC’) and peak acute fall in score (worst score) were computed for each patient. IBDQ-B AUC was calculated by firstly subtracting scores from the maximum attainable: 70. The transformed scores were then summed as: [0.5 * baseline + sum of interim acute scores + 0.5 * final acute score]. Over a 6 week course of radiotherapy AUC scores could thus range from 0 to 360 with a lower AUC value indicating lesser toxicity.
2.6 Statins and ACE inhibitor use
A proforma was designed to capture data from the patients’ General Practitioners (GPs) on brand, dose and duration of statin and/or ACEi usage during radiotherapy treatment (dates pre-entered on the proforma by a study investigator) and at 1 year post treatment. Proformas were posted (surface mail) or faxed to each GP with an accompanying letter and copy of the published abstract
of the original study in which their patient participated. GPs who did not respond initially were contacted for a second final time.
2.7 Statistical considerations and analysis
This study was not powered to a specific endpoint. Means and standard deviations for non-normally distributed data were computed for change in IBDQ-B scores by study cohort and by statin and/or ACEi subgroups: ‘statin users’; ‘non-statin users’; ‘ACEi users’; ‘non-ACEi users’; ‘statin + ACEi users’; ‘non-statin + ACEi’ users. Worst IBDQ-B and IBDQ-B AUC scores were also computed. IBDQ-B AUC scores were transformed using a natural logarithm transformation to correct for non-normality of data. An analysis of variance (ANOVA) was planned to explore the possible association between acute IBDQ-B AUC (previously shown to be a dose-independent predictor of late toxicity)
A total of 308 patients (originally enrolled between January 2004 and October 2006) were eligible for inclusion in this extended analysis. Ten were excluded due to missing consent forms. Patients’ GPs were contacted between June 2010 and February 2011. Of 298 GPs contacted, 253 responded (84%). Two hundred and thirty-seven patients had evaluable acute toxicity and statin or ACEi data with 51% having no missing IBDQ-B scores. In the remaining patients, 70% (n = 116) were missing one score, 24% (n = 56) 2–4 scores and 6% (n = 14) ⩾5 scores. At 1 year, 164 patients had 1 year follow-up data (Fig. 1). Of the n = 73 not available for follow up at 1 year, 43 had died, 27 had moved away, 18 declined follow-up, 2 were too poorly to respond, 1 withdrew consent and in 3 patients no reason was recorded.
Fig. 1CONSORT style analysis of patient/data accrual.
Patient characteristics for the 308 patients with evaluable acute data are given in Table 2. For all subgroups ‘users’ were older than ‘non-users’ (73.5 versus 67 years; 72 versus 68 years; 72 versus 69 years in statin (p = 0.002), ACEi (p = 0.111), and statin + ACEi users respectively (p = 0.381) and in receipt of higher median radiotherapy dose (64 versus 55.8 Gy; 60 versus 55.8 Gy; 64 versus 55.8 Gy in statin (p = 0.014), ACEi (p = 0.315) and statin + ACEi (p = 0.276) users respectively. Males outnumber females in user groups by 3-fold (statin users), 2-fold (ACEi users) and 2-fold (statin + ACEi users). Males comprised 57% of the entire cohort of patients. At 1 year, the 164 patients with evaluable data comprised GI: 15% (n = 24), Gy: 34% (n = 56), U: 51% (n = 84) patients. The proportion of males had risen to 62% (male:female ratio = 101:63). Mean (sd) age: 68.7 (11.07) years and median (range) radiotherapy dose 55.8 Gy (20–74).
Table 2Baseline characteristics.
Characteristic
Statin use
ACE inhibitor use
Statins + ACE inhibitor use
Total
Users
Non-users
Users
Non-users
Users
Non-users
n (%)
38 (16)
199 (84)
39 (16.5)
198 (83.5)
18 (7.6)
219 (92.4)
237
Gender – male:female
29:9
106:93
25:14
110:88
12:6
123:96
135:102
Age, years: median (range)
73.5 (59–86)
67 (29–88)
72 (55–83)
68 (29–88)
72 (59–81)
69 (29–88)
69 (29–88)
Dose, Gy: median (range)
64 (36–74)
55.8 (20–74)
60 (45–74)
55.8 (20–74)
64 (45–70)
55.8 (20–74)
55.8 (20–74)
Pelvic sites and use of concomitant chemotherapy: n (%)
Acutely (i.e. at baseline and during radiotherapy) n = 38 patients (16%) were taking statins: atorvastatin n = 18; simvastatin n = 14; pravastatin n = 5; rosuvastatin n = 1. Dose ranges were reported by GPs as: atorvastatin 10–80 mg od; simvastatin 20–40 mg od; pravastatin 20 mg od; rosuvastatin 10 mg od. A total of n = 39 patients (16.5%) were taking ACEi: ramipril n = 14; lisinopril n = 10; perindopril n = 8; fosinopril n = 4. Dose ranges were reported by GPs as: ramipril 2.5–10 mg od; lisinopril 2.5–20 mg od; perindopril 2–8 mg od; fosinopril 10–20 mg od. Acutely, n = 18 patients (7.6%) were taking statin + ACEi in the following combinations: atorvastatin + ACEi n = 10; simvastatin + ACEi n = 5; pravastatin + ACEi n = 3, the most common combination being atorvastatin + ramipril in n = 5 patients. At 1 year, n = 26 patients (16%) were still taking statins (missing: 12), n = 27 (16%) were still taking ACEi (missing: 12) and n = 14 (8.5%) patients remained on statins + ACEi agents (missing; 4).
3.4 Gastrointestinal toxicity – acute setting
3.4.1 Entire cohort
Change in IBDQ-B scores acutely (n = 237) is given in Table 3 and depicted in Fig. 2. IBDQ-B scores fell by 6.9 points from mean (sd) 66.0 (7.5) at start of radiotherapy to 59.1 points (10.0) at end of radiotherapy (n = 4 missing at end-RT).
Mean (sd) changes in acute IBDQ-B scores by subgroups are given in Table 3 and depicted in Fig. 2. The mean changes in score in non-user subgroups between the start and end of radiotherapy were 7.3 points (non-statin users), 7.3 (non-ACEi users) and 7.0 (non-statin + ACEi users). In comparison, mean changes in score in user sub-groups were 4.8 points (statin users), 5.0 points (ACEi users) and 4.9 points (statin + ACEi users). Baseline IBDQ-B scores were higher in statin users compared to non-statin users by 2.6 points, in ACEi users (compared to non-users) by 1.2 points and in statin + ACEi users (compared to non-users) by 3.4 points.
Table 4 shows the worst acute IBDQ-B score and IBDQ-B AUC by subgroup including the difference in score between users and non-users in each subgroup. User subgroups experienced less severe peaks in acute toxicity (worst IBDQ-B) than non-user sub-groups and reduced overall acute severity of toxicity (lesser IBDQ-B AUC scores). The difference in scores was greatest between statin + ACEi users compared to non-statin + ACEi users, with a difference of 7.4 points for worst acute IBDQ-B score and 30 for IBDQ-B AUC score.
Analysis of variance showed statin use to be significantly associated with acute IBDQ-B AUC (p = 0.04) after controlling for baseline scores whilst ACEi use was not (p = 0.20). Combined statin + ACEi use was more strongly associated with IBDQ-B AUC (p = 0.008) after correcting for baseline scores.
3.4.3 Within subgroups
Within subgroups (defined by individual agent) acute mean (sd) fall in IBDQ-B score (worsening toxicity) was least amongst pravastatin users, ranging from a fall of 6.22 (8.27) points for atorvastatin to 2.40 (4.51) for pravastatin users. Acute mean (sd) fall in IBDQ-B score in the ACEi users was least amongst perindopril users, ranging from a fall of 18.0 (6.93) for fosinopril to 0.1 (10.37) points for perindopril users. In the combined agent subgroup, acute mean (sd) fall in IBDQ-B score ranged from 5.8 (5.84) points for simvastatin + ACEi users to 3.3 (6.11) points for pravastatin + ACEi users.
3.5 Gastrointestinal toxicity – 1 year
At 1 year, mean (sd) IBDQ-B scores (n = 164) had fallen 2.8 points – Table 3. By subgroups, mean IBDQ-B scores in non-user groups had fallen by 2.7 points in non-statin users (n = 138), 3.1 points in non-ACEi users (n = 137) and 2.6 points in non-statin + ACEi users (n = 150). In user sub-groups scores had fallen 3.0 points in statin users (n = 26), 1.1 points in ACEi users (n = 27) and 4.0 points in statin + ACEi users (n = 14). Mean (sd) IBDQ-B scores at 1 year remained higher in all users than non-user subgroups – Table 3. Non-normality of data precluded further formal analysis.
Within subgroups, mean (sd) fall in 1 year IBDQ-B scores was similar amongst statin users ranging from 3.5 (4.34) points for atorvastatin users to 4.1 (4.87) points for simvastatin users. The 1 rosuvastatin user increased their 1 year score by 1 point. Amongst ACEi users, mean (sd) fall in score ranged from 1.7 (3.51) points for fosinopril users to 9.6 (11.0) points for perindopril users. However, in enalopril and lisinopril users, scores had risen by mean (sd) 0.5 (3.54) and 3.0 (6.51) points respectively. Within the combined agent subgroup, acute mean (sd) fall in IBDQ-B scores was very similar: 3.0 (2.94) points for simvastatin + ACEi users, 4.0 (5.29) points for pravastatin + ACEi users and 4.0 (4.18) for atorvastatin + ACEi users.
3.6 Survival outcomes
A total of 43 patients died before their 1 year follow-up. Of these, 26 died of disease progression, (of whom n = 20 were non-users and n = 6 users) and 7 from non-cancer related causes. Cancer related treatment toxicity (i.e. chemotherapy and/or radiotherapy and/or surgery) significantly contributed to morbidity before death in the remaining 10 patients of whom n = 7 were non-users and n = 3 users.
4. Discussion
This study has demonstrated for the first time that the use of statin or statin + ACEi during pelvic radical radiotherapy significantly reduces gastrointestinal toxicity, assessed using a gastrointestinal scoring tool (IBDQ-B) that we have previously shown to be more sensitive than other commonly used tools in this setting
A modified inflammatory bowel disease questionnaire and the Vaizey incontinence questionnaire are more sensitive measures of acute gastrointestinal toxicity during pelvic radiotherapy than RTOG grading.
We conducted this investigation in a mixed pelvic cohort without stratification by pelvic site or disease stage and obtained data on medication use retrospectively. It was also possible that the number of men compared to women in each user group, and the type of radiotherapy they received may have influenced our findings. We were reluctant to perform additional unplanned statistical analyses in our data, however, an exploratory ANOVA revealed that both statin and statin + ACEi use remained significantly associated with IBDQ-B AUC (p < 0.001) after controlling for gender (data not shown) suggesting that our findings are indeed valid.
A further criticism by those not familiar with the IBDQ-B scoring system used in this study could be that the changes in scores between users and non-users seem modest, however, changes in scores of >6 points have previously been shown to be clinically significant.
We were able to obtain this data from patient’s GPs. This data has enabled us to observe that pravastatin may be providing enhanced protection not only in comparison to other statins but also in combination with ACEi agents. Although the numbers of patients on these particular agents is very small (pravastatin: 5 and pravastatin + ACEi: 3), there is sound biological reasoning to suggest that this non-lipophilic statin may have an enhanced effect
as it evades first pass metabolism by the liver and has been shown to concentrate >300 times more than other (lipophilic) statins in peripheral tissues. It may also have localised effects in the region of the terminal ileum and distal small bowel being selectively taken-up by the sodium-independent bile acid transporter.
In many of the previous studies on the use of medication concomitant with radiotherapy, medication use has been largely assessed by a clinician on the basis of patient’s self-reported history or surmised from presence of the co-morbidity.
Clinical factors predicting late severe urinary toxicity after postoperative radiotherapy for prostate carcinoma: a single-institute analysis of 742 patients.
35% of the patients in the current analysis (82/237) said they were on treatment for hypertension. However, information from the same patients’ GPs revealed that in fact n = 39 were on ACEi agents leaving 52% on alternative HT medication thus illustrating the need for precise data when drawing conclusions on the use of individual agents and effects.
It could be argued that in this investigation, all user groups were less symptomatic on presentation than non-users and that this may have influenced outcomes. However, the significant association between IBDQ-B AUC and statin use and IBDQ-B AUC statin + ACEi use was ascertained using an analysis of variance which accounted for baseline values. The phenomena of statin users being in better health than their non-user counterparts has been noticed previously
and emphasises the importance of collecting detailed baseline characteristics. It is possible that statin users have heightened health awareness and so present earlier in the disease cycle. As we have also reported in this analysis, statin users (despite being older) are prescribed higher median radiotherapy doses and thus may also benefit with improved survival and biochemical outcomes. On the other hand, both older age and increased dose are risk factors for increased toxicity. Prospective investigation stratified by disease stage and agent/dose with adequate follow-up will help to answer these observations.
We conclude that both statin and ACEi agents (or a combination of these agents) may confer multiple advantages in the therapeutic pelvic radiotherapy setting. Adequately powered investigations with meticulous detail regarding medication use will help to unravel the precise effects of these potentially protective agents.
Conflict of interest statement
None declared.
Acknowledgements
We would like to thank all General Practitioners who responded to our request for information, without whose help this analysis would have been impossible.
This work was undertaken in The Royal Marsden NHS Foundation Trust who received a proportion of its funding from the NHS Executive; the views expressed in this publication are those of the authors and not necessarily those of the NHS Executive. This work was supported by the Institute of Cancer Research, The Bob Champion Cancer Trust and Cancer Research UK Section of Radiotherapy [CUK] Grant No. C46/A2131. We acknowledge NHS funding to the NIHR Biomedical Research Centre.
References
Kollmeier M.A.
Katz M.S.
Mak K.
et al.
Improved biochemical outcomes with statin use in patients with high-risk localised prostate cancer treated with radiotherapy.
Clinical factors predicting late severe urinary toxicity after postoperative radiotherapy for prostate carcinoma: a single-institute analysis of 742 patients.
A modified inflammatory bowel disease questionnaire and the Vaizey incontinence questionnaire are more sensitive measures of acute gastrointestinal toxicity during pelvic radiotherapy than RTOG grading.
00099-8&id=gr1.jpg){kind=link}
00099-8&id=gr2.jpg){kind=link}