Pancreatic ductal adenocarcinoma (PDAC) is by far the most common form of pancreatic cancer, accounting for ~90% of cases, and possesses a notoriously poor prognosis; overall only 21% of patients survive for one year or more, and just 3% for 5 years or more . Much of the previous literature suggest that it has often metastasized at the time of presentation, which is reflected by the fact that even with small cancers (<1 cm) 30% have node-positive disease and 10% have distal metastasis [2,3].
Gemcitabine and the combination therapy FOLFIRINOX are the two principal chemotherapeutic agents for PDAC. These therapeutic options have been shown to deliver modest improvements of overall survival, but are associated with significant toxicity and an often unacceptable side effect profile; attenuated doses of FOLFIRINOX are frequently required as patients are unable to tolerate it and they are often elderly .
Over recent years, there has been increasing interest in repurposing nonanticancer drugs for use in oncology [5,6]. This has ‘substantial advantages’ over de-novo discovery of anticancer drugs including; an established safety profile, faster progression into phase 3 clinical trials and more cost-effective treatments .
Calcium signalling has been implicated in both tumourigenesis and tumour progression, with recent literature reviews highlighting the role of calcium channels in; apoptosis evasion, abnormal proliferation, abnormal differentiation, invasion and tumour metastasis [7,8]. Calcium channel blockers (CCBs) are a widely prescribed class of drugs used in many common conditions such as hypertension, angina and atrial fibrillation. Multiple basic science studies have postulated a therapeutic potential of CCBs in cancer [9–18]. They are well tolerated, inexpensive, and have a favourable side effect profile when compared to standard anticancer drugs. This could make CCBs ideal candidates for drug repurposing into viable adjuncts to currently available chemotherapy regimens.
The widespread usage of CCBs affords an opportunity to investigate whether their hypothesized benefit translates into the clinical setting by retrospectively studying coincidental CCB prescription and overall survival in PDAC patients. Our previous retrospective cohort study demonstrated that patients prescribed CCBs displayed significantly improved overall survival following pancreaticoduodenectomy for PDAC (hazard ratio 0.475, P = 0.023); 26.8 months for those prescribed CCBs versus 17.4 months for those not taking the drug . This study aimed to further evaluate the effect of CCBs on overall survival in PDAC patients, with a focus on an unresectable cohort.
All patients included in the study were adults with unresectable [American Joint Committee on Cancer (AJCC) Stage 3/4] histologically confirmed PDAC. Included patients all attended an appointment with an oncologist to discuss chemotherapy, and were eligible for inclusion regardless of their treatment decision. Patients with a ‘curative’ resection, AJCC stage 1 or 2 disease, active nonpancreatic malignancy and patients without a biopsy diagnosis were excluded. All patients were treated in the UK, and remained in the UK throughout follow-up. Information was retrospectively collected from electronic and hard copy patient notes. Data were extracted on drug history, and for a range of other potentially important prognostic indicators: Eastern Cooperative Oncology Group (ECOG) performance status, AJCC stage (stage 3 = unresectable due to local invasion, stage 4 = presence of metastasis), chemotherapy regimen, radiotherapy, patient age, hypertensive status and sex. Staging data was based on initial staging computed tomography scan. Details on ethnicity could unfortunately not be reliably ascertained from patient notes. Overall survival from biopsy diagnosis was selected as our study outcome.
The study was registered with our local research governance department and an appropriate Caldicott application was completed. Informed consent was not required.
The Kaplan–Meier method was used to calculate estimated median overall survival, with Logrank tests applied to compare groups. Censoring was applied, allowing any patients who were alive at the end of follow-up to be included in the analysis. The Pearson Chi-Square test was used to compare categorical variables and independent two-tails t-tests were used to compare means.
A Cox proportional hazards model was used to compare overall survival between groups, adjusting for potential confounders; ECOG performance status (≤1 vs >1), AJCC stage (3 vs 4; i.e., presence of metastasis), chemotherapy regimen (none vs gemcitabine based vs FOLFIRINOX), radiotherapy, hypertensive status, sex and age (≤60 vs >60).
IBM SPSS Statistics was used for statistical analysis. Graphs were generated using MedCalc version 18.9 statistical analysis software. An alpha level of 0.05 was selected.
In total, 164 patients with unresectable histologically confirmed pancreatic ductal adenocarcinoma were referred to the Northern Centre for Cancer Care between October 2010 and January 2017. Sufficient data were available to allow inclusion of all patients. Median follow-up time was 10.0 months; until death in 149 patients.
A total of 30 patients were taking CCBs at the time of biopsy diagnosis. Of these 30 patients, 22 were taking amlodipine (5–10 mg), four nifedipine (40 mg), two lercanidipine (10 mg), one felodipine (5 mg) and one patient was taking diltiazem (180 mg). The indication for CCB prescription was hypertension in 28 patients, and angina in the remaining two patients.
Of the 164 participants, 74 received gemcitabine-based chemotherapy regimens, 62 received FOLFIRINOX and 26 participants did not receive chemotherapy. Patients classified as receiving gemcitabine-based regimens received either gemcitabine alone (n = 60), or gemcitabine combined with: Abraxane (n = 5), Capecitabine (n = 6), vandetanib (n = 2) or cisplatin (n = 1).
Table 1 displays the comparison of baseline demographics between the ‘prescribed CCB’ and ‘no CCB prescribed’ groups. Participants taking CCBs were older (P = 0.001) and more likely to have hypertension (P < 0.0005). Other baseline demographic factors, including chemotherapy regimen, receipt of radiotherapy and diabetes status showed no significant difference between groups.
Kaplan–Meier survival curves can be found in Fig. 1. Overall, the Kaplan–Meier estimated median survival was 15.3 months for patients prescribed CCBs vs 10.1 months for patients not prescribed CCBs (Fig. 1a; P = 0.131, Logrank). For participants who received FOLFIRINOX (n = 62), estimated median survival was 20.8 months for those prescribed CCBs (n = 9) vs 12.8 months for the remainder (n = 53); P = 0.088, Logrank (Fig. 1b). For participants receiving gemcitabine-based chemotherapy (n = 76), estimated median survival was 11.2 vs 10.4 months in participants prescribed versus not prescribed CCBs, respectively (Fig. 1c; P = 0.580, Logrank).
Table 2 shows the results of our Cox proportional hazards models. ECOG performance status, AJCC cancer stage, radiotherapy and chemotherapy regimen were all predictors of survival on simple, unadjusted Cox regression (P < 0.0005 for all four factors). When adjusting for the potential confounders listed in Table 2, participants prescribed CCBs demonstrated significantly improved overall survival compared those without CCBs prescribed; hazard ratio 0.496 (95% CI = 0.297–0.827; P = 0.007). ECOG performance status (P < 0.0005), AJCC cancer stage (P < 0.0005), participant age (P = 0.012), radiotherapy (P = 0.002) and chemotherapy regimen (P < 0.0005) remained significant predictors of survival in the adjusted Cox model.
Previous work had suggested that the reason for the improved survival in the CCB group may be due to a subgroup who was also taking aspirin. In the present cohort, six of 164 patients were taking both aspirin and CCBs in combination. There was no evidence on simple (hazard ratio 0.920, P = 0.843) or adjusted cox regression (hazard ratio 0.822, P = 0.644) that patients taking both drugs in combination had a survival benefit over the rest of the cohort.
This retrospective cohort study looked at the effect of CCBs on overall survival in patients with unresectable PDAC. Adjusted Cox regression revealed significantly improved overall survival in patients prescribed CCBs; hazard ratio 0.496 (95% CI = 0.297–0.827; P = 0.007). The Kaplan–Meier estimated median survival was longer for patients prescribed CCBs (15.3 vs 10.1 months), but this did not reach statistical significance (P = 0.131).
The improved survival demonstrated by adjusted Cox regression (hazard ratio 0.587) was similar to that seen in our previous study looking at the effect of CCBs in patients undergoing a Whipple’s resection for PDAC (hazard ratio 0.475, P = 0.023), where overall survival was 26.8 months for those prescribed CCBs vs 17.4 months for those not taking the drug .
Repurposing of noncancer medications for use in oncology has ‘substantial advantages’ over the de-novo development of anticancer drugs . CCBs are ideal candidates for repurposing as they are commonly prescribed, with limited side effects, and previous laboratory studies suggest anticancer properties. This significant body of existing literature that offers potential mechanisms by which CCBs may have a therapeutic benefit in cancer [7,8]. Potential effects include promoting apoptosis, inhibiting proliferation and metastasis and sensitizing PDAC cells to chemotherapy.
It has long been recognized that intracellular calcium signalling is integral to apoptosis, the highly regulated process of cell death which is often impaired within cancer cell populations . Various CCBs have been shown to increase apoptosis in transformed cell lines including glioblastoma and colon adenocarcinoma in vitro [9,10]. Interestingly, these studies also demonstrated that CCBs can potentiate the apoptotic effects of traditional chemotherapeutic agents, suggesting a possible chemosensitizing effect. A more recent in-vitro study using pancreatic cancer cell lines has found that CCBs alone can induce apoptosis in PDAC.
In addition to promoting apoptosis, there is evidence that CCBs can ameliorate the abnormal proliferation seen in cancerous cells. Two studies demonstrate that CCBs can decrease tumour growth and inhibit proliferation in vivo [12,13]. One of these injected HT-39 breast cancer cells into an athymic mouse, and found that the CCBs amlodipine, verapamil and diltiazem were all able to inhibit proliferation and lead to significant decreases in tumour growth . Uehara et al.  used a rat model with N-nitrosomorpholine to induce HCC formation and found that treatment with CCB led to less cancer formation, and decreased proliferation where cancers did form. Especially, pertinent to the present study are three independent in-vitro studies showing that CCBs are able to decrease proliferation and inhibit growth, specifically in pancreatic cancer cell lines [11,14,15]. Importantly, this effect was seen with a variety of CCBs in numerous pancreatic cancer cell lines.
Others have described calcium signalling as a crucial regulator of tumourigenic cell migration and metastasis . In-vivo models have demonstrated the ability of CCBs to prevent metastasis of lymphosarcoma, melanoma and colon adenocarcinoma, theorizing that inhibition of platelet activation as a possible underlying mechanism [16,17]. Recent work with PDAC cell lines has found that CCBs reduce migration and invasion, suggesting that this antimetastatic effect applies to pancreatic cancer .
Another potential mechanism for the benefits of CCB seen in the present study is the ability to sensitize cancer cells to chemotherapy, as discussed above [9,10]. One of the main reasons is the ability of CCBs to inhibit multidrug resistance protein 1 (MDR1, also known as p-glycoprotein) . Timcheva and Todorov  showed that CCBs were able to potentiate the effects of chemotherapy on leukaemia cell lines that are high expressers of MDR1, but not leukaemia cell lines which were low expressers. CCBs ability to block MDR1 may be particularly important in PDAC, a cancer which is notoriously chemoresistant, and which is also a high expresser of MDR1 .
Calcium channels have been implicated in a wide variety of tumourigenic processes, including several hallmarks of cancer; resisting apoptosis, sustaining proliferative signalling, activating invasion and metastasis . The basic science papers referenced above represent a large body of research to support the hypothesis that blockade of calcium channels may have a therapeutic benefit in patients with cancer.
The above basic science studies use a range of different types of CCBs. While these all have similar effects, their chemical structure differs and they have slightly different pharmacological properties. While we did collect data on which type of CCB patients in this study were taking, the sample size was too small to compare survival differences between different types of CCBs. Future work using cancer registries could achieve a far larger sample size and allow comparison between different types of CCB at different doses.
The main limitation of this study is the retrospective design, which prevents definitive conclusions from being drawn. This leads to a selection bias; patients taking CCBs are older, more comorbid and likely have a poorer functional status. This may be the reason for the nonsignificant survival differences on the Kaplan–Meier analysis and unadjusted Cox regression; only when this selection bias was combatted by adjusting for these confounders did the improved survival in the CCB group reach significance.
Another limitation is the fact that data on prescription is based on medications at discharge from hospital following diagnostic biopsy. It was not possible to reliably ascertain whether patients had any adverse reactions to CCB. As patients may have subsequently stopped taking CCBs, any survival benefits may be underestimated by our study.
In conclusion, this study supports previous findings that CCBs may prolong survival in pancreatic cancer. While current treatments for pancreatic cancer are toxic and come with a multitude of severe side effects, CCBs are widely prescribed, well tolerated, inexpensive drugs with minimal side effects. These properties would make it feasible for CCBs to be repurposed as an adjunct to currently available chemotherapy regimens.
Future work using large, multicentre databases would allow more powerful analyses, including the ability to further investigate the effects of different types of CCBs, at different doses and in combination with different chemotherapy regimens. This data could then inform the design of a randomized controlled trial which is essential to evaluate the true effect of CCBs in pancreatic cancer.
Conflicts of interest
There are no conflicts of interest.
2. Stathis A, Moore MJ. Advanced pancreatic carcinoma: current treatment and future challenges. Nat Rev Clin Oncol. 2010; 7:163–172
3. Yu J, Blackford AL, Dal Molin M, Wolfgang CL, Goggins M. Time to progression of pancreatic ductal adenocarcinoma
from low-to-high tumour stages. Gut. 2015; 64:1783–1789
4. Conroy T, Desseigne F, Ychou M, Bouché O, Guimbaud R, Bécouarn Y, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer
. N Engl J Med. 2011; 364:1817–25
5. Bertolini F, Sukhatme VP, Bouche G. Drug repurposing
in oncology – patient and health systems opportunities. Nat Rev Clin Oncol. 2015; 12:732–742
6. Pushpakom S, Iorio F, Eyers PA, Escott KJ, Hopper S, Wells A, et al. Drug repurposing
: progress, challenges and recommendations. Nat Rev Drug Discov. 2019; 18:41–58
7. Martínez-Delgado G, Felix R. Emerging role of CaV1.2 channels in proliferation and migration in distinct cancer cell lines. Oncology. 2017; 93:1–10
8. Prevarskaya N, Skryma R, Shuba Y. Calcium in tumour metastasis: new roles for known actors. Nat Rev Cancer. 2011; 11:609–618
9. Kondo S, Yin D, Morimura T, Takeuchi J. Combination therapy with cisplatin and nifedipine inducing apoptosis in multidrug-resistant human glioblastoma cells. J Neurosurg. 1995; 82:469–474
10. Shchepotin IB, Soldatenkov V, Buras RR, Nauta RJ, Shabahang M, Evans SR. Apoptosis of human primary and metastatic colon adenocarcinoma cell lines in vitro
induced by 5-fluorouracil, verapamil, and hyperthermia. Anticancer Res. 1994; 14:1027–1031
11. Woods N, Trevino J, Coppola D, Chellappan S, Yang S, Padmanabhan J. Fendiline inhibits proliferation and invasion of pancreatic cancer
cells by interfering with ADAM10 activation and β-catenin signaling. Oncotarget. 2015; 6:35931–35948
12. Taylor JM, Simpson RU. Inhibition of cancer cell growth by calcium channel antagonists in the athymic mouse. Cancer Res. 1992; 52:2413–2418
13. Uehara H, Nakaizumi A, Baba M, Iishi H, Tatsuta M. Inhibition by verapamil of hepatocarcinogenesis induced by N-nitrosomorpholine in Sprague-Dawley rats. Br J Cancer. 1993; 68:37–40
14. Sato K, Ishizuka J, Cooper CW, Chung DH, Tsuchiya T, Uchida T, et al. Inhibitory effect of calcium channel blockers
on growth of pancreatic cancer
cells. Pancreas. 1994; 9:193–202
15. Jäger H, Dreker T, Buck A, Giehl K, Gress T, Grissmer S. Blockage of intermediate-conductance Ca2+-activated K+ channels inhibit human pancreatic cancer
cell growth in vitro
. Mol Pharmacol. 2004; 65:630–638
16. Tsuruo T, Iida H, Makishima F, Yamori T, Kawabata H, Tsukagoshi S, Sakurai Y. Inhibition of spontaneous and experimental tumor metastasis by the calcium antagonist verapamil. Cancer Chemother Pharmacol. 1985; 14:30–33
17. Jurásková V, Sládek T. Antimetastatic action of diltiazem on LS/BL tumor cells in liver tumor-colony assay. Neoplasma. 1990; 37:343–348
18. Timcheva CV, Todorov DK. Does verapamil help overcome multidrug resistance in tumor cell lines and cancer patients? J Chemother. 1996; 8:295–299
19. Tingle SJ, Moir JA, White SA. Role of anti-stromal polypharmacy in increasing survival after pancreaticoduodenectomy for pancreatic ductal adenocarcinoma
. World J Gastrointest Pathophysiol. 2015; 6:235–242
20. Mason RP. Effects of calcium channel blockers
on cellular apoptosis: implications for carcinogenic potential. Cancer. 1999; 85:2093–2102
21. Goldstein LJ. MDR1 gene expression in solid tumours. Eur J Cancer. 1996; 32A:1039–1050
22. O’Driscoll L, Walsh N, Larkin A, Ballot J, Ooi WS, Gullo G, et al. MDR1/P-glycoprotein and MRP-1 drug efflux pumps in pancreatic carcinoma. Anticancer Res. 2007; 27:2115–2120
23. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011; 144:646–674