Survival After Solid Cancers in Antithrombotic Trials




The impact of antithrombotics on cancer is currently under intense investigation because of the excess of solid cancers in trials after thienopyridines such as TRITON (prasugrel), DAPT (prasugrel and clopidogrel), PAR-1 thrombin antagonist in TRACER (vorapaxar), pyrimidines in PEGASUS (ticagrelor), and in APPRAISE-2 after apixaban. However, whether patient survival after solid cancer (SASC) in antithrombotic trials may be affected is unknown. We matched the 1-year SASC rate in antithrombotic trials reported by Food and Drug Administration with the census averages in Surveillance, Epidemiology, and End Results (SEER) Program by the US National Cancer Institute and World Health Organization (WHO) surveys. The Food and Drug Administration provided the SASC data for 3 trials with similar cancer survival of about 70% for the first year of follow-up in TRITON, APPRAISE-2, and ARISTOTEL. Adjusted cancers in TRITON with SEER (odds ratio 0.92; 95% confidence interval 0.53 to 1.59, p = 0.4351) and WHO (odds ratio 0.99; 95% confidence interval 0.57 to 1.7, p = 1.00) revealed very close if not identical SASC rates in antithrombotic trials compared to epidemiologic census estimates. In conclusion, SASC rates in patients enrolled in antithrombotic trials do not differ from SEER or World Health Organization averages.


The link between malignancy and hemostasis is not new. The first alarming signal that antiplatelet therapy may cause excess of solid cancers was observed in the prasugrel arm of TRITON-TIMI 38 trial and confirmed by the Food and Drug Administration (FDA). Later, until TRACER FDA review, the data remain inconclusive, probably because of the awareness of drug manufacturers of such catastrophic association resulted in missing delayed outcomes, discontinuations, or/and incomplete follow-up, suggesting bias. Recently, however, 4 important pieces of the puzzle emerge, namely vorapaxar, apixaban FDA reviews, DAPT for clopidogrel and prasugrel, and PEGASUS for ticagrelor confirming that cancer signal after antithrombotics is not an artifact, or play of chance, but rather a scaring reality. Independently, a Danish registry for almost 126,000 myocardial infarction (MI) survivors without a history of cancer were matched with more than 3 million subjects with no history of either MI or cancer. Results suggested that cancer rate was more than doubled (173.5 per 10,000 person-years) in the MI survivors versus 85.2 per 10,000 person-years for their peers without MI over 17-year follow-up. Despite the existing cancer signal, survival of such patients may be worsened because chronic antithrombotic regimens may cause not only more cancer dissemination but promotion of more aggressive unclassified malignancies associated with potentially shorter survival. We, therefore, matched the survival after solid cancers in TRITON prasugrel arm with the publically available evidence from 2 reputable independent international census datasets.


Methods


The cancer data for TRITON are readily available for prasugrel New Drug Application review and updated in several later public FDA documents. Two data sets were used as the references. We matched the TRITON evidence with the Surveillance, Epidemiology, and End Results (SEER) Program, issued by the US National Cancer Institute and corresponding World Health Organization (WHO) data gathering survival data from 14 regions worldwide. For survival comparisons, SEER Survival Database (SEER 13) from 1988 until 2010 has been used adjusting age (>50 years), race (white and Hispanic), gender (both), and corresponding site of cancer location. For incidence data, Age-Adjusted SEER Incidence Rates (SEER 9) with selected year 2009, and similar age, race, gender, and cancer location adjustments were used. Analyses were done with STATA 13 software (StataCorp, College Station, Texas) using unstratified cumulative incidence data function with Fisher’s exact test with computation of 95% confidence interval (CI) and odds ratios (ORs).




Results


The distribution of 132 solid cancers in prasugrel arm of TRITON trial is presented in Figure 1 . The sites for solid cancers in TRITON were broad, and the most prevalent locations were colorectal, lung, prostate, urinary bladder, and stomach. The solid cancer survival curves from the TRITON FDA review are presented in Figure 2 . In fact, cancer survival in TRITON did not differ much dependent on treatment assignment and was overall about 70% for 1 year in both clopidogrel and prasugrel arms. Importantly, most registries and large data sets report cancer survival in 5-year increments, representing some difficulties in adjusting the trial data, in which the follow-up is usually much shorter and sometimes is even less than 1 year. The 5-year survival rates for new solid cancers based on the SEERSTAT US National Cancer Institute data are presented in Table 1 . Not surprisingly, the US survival rates were better than worldwide estimates because the latest include the patients from underdeveloped countries. Because most of the patients in antithrombotic trials require routine cardiac catheterization, and, therefore, enrolled in established top medical centers, TRITON cancer survival should be matched much more reasonably with the US-generated data sets. The differences between actual solid cancers in TRITON after prasugrel and SEER’s modeled data for cancer sites are presented in Table 2 . The incidence of new solid cancers in TRITON overall was similar to those in SEER; however, in TRITON, there were significantly more colorectal and stomach and less prostate cancers. Importantly, SEER statistics compare the survival of patients diagnosed with cancer with the survival of people in the general population who are the same age, race, and gender and who have not been diagnosed with cancer. As a limitation, survival statistics are based on large groups of people, they cannot be used to predict exactly what will happen to an individual patient because no 2 patients are entirely alike, and treatment and responses to treatment can vary greatly. Adjusted cancer survival rates observed in TRITON- and SEER-modeled data for cancer locations are presented in Table 3 .




Figure 1


Distribution of new cancer locations in TRITON.



Figure 2


Survival after first solid cancer events in TRITON (A) , ARISTOTELE (B) , and APPRAISE-2 (C) .


Table 1

National cancer institute 5-year survival rates for new solid cancers




















































Cancer Location 5-Years Survival US 5-Years Survival Worldwide (range)
Prostate 98.9% 48-99%
Colorectal 64.7% 39-65%
Breast 89.2% 73-89%
Urinary Bladder 77.4% 39-78%
Kidney 72.4% 45-72%
Pancreas 6.7% 3-10%
Ovarian 44.6% 30-45%
Stomach 28.3% 15-30%
Lung 16.8% 8-17%
Esophagus 17.5% 5-19%
Oral/pharynx 62.7% 48-69%


Table 2

SEER’s adjusted incidence of cancer locations in TRITON












































































Cancer Site Solid Cancers Incidence
(SEER)
Incidence
(TRITON)
Odds ratio with 95% CI; and p-value
Prostate 17 524.43 32.38 0.530 (0.296 to 0.947, p = 0.044)
Colorectal 29 134.93 11.11 2.643 (1.337 to 5.226, p = 0.006)
Breast 4 364.93 7.51 0.499 (0.160 to 1.560, p = 0.387)
Urinary Bladder 13 78.19 6.44 2.169 (0.853 to 5.518, p = 0.167)
Kidney 9 45.85 3.77 2.252 (0.735 to 6.893, p = 0.267)
Pancreas 5 43.46 3.58 1.250 (0.364 to 4.301, p = 1.000)
Ovarian 3 38.17 0.79 3.001 (0.348 to 25.855, p = 0.625)
Stomach 14 20.18 1.66 7.012 (1.966 to 25.014, p = 0.004)
Lung 28 208.72 17.18 1.650 (0.910 to 2.990, p = 0.135)
Esophagus 7 16.72 1.38 7.005 (1.160 to 42.306, p = 0.07)
Oral/pharynx 3 33.87 2.79 Identical, not significant

Solid cancers in prasugrel arm of TRITON trial, with the median follow-up of 14.5 months.


Per 100,000 persons per 1 year, year 2009 ages 50+, whites or Hispanics, both sexes or as appropriate females (breast, ovarian) or males (prostate) via (SEER 9).


TRITON rate modeled by SEER database.



Table 3

Adjusted cancer survival in TRITON and SEER-modeled data





































































Cancer Site Solid Cancers in TRITON
(n, %)
Relative Survival at 1 Year SEER TRITON Mortality at 1 Year Approximated from SEER
Prostate 17 (12.9%) 99.7% 0
Colorectal 29 (22.0%) 81.6% 5
Breast 4 (3.0%) 97.5% 0
Urinary Bladder 13 (9.8%) 90.1% 1
Kidney 9 (6.8%) 79.0% 2
Pancreas 5 (3.8%) 22.4% 4
Ovarian 3 (2.2%) 70.8% 1
Stomach 14 (9.0%) 46.7% 7
Lung 28 (21.3%) 40.7% 17
Esophagus 7 (5.2%) 44.8% 4
Oral/pharynx 3 (2.2%) 82.0% 1
Sum 132 42

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Nov 28, 2016 | Posted by in CARDIOLOGY | Comments Off on Survival After Solid Cancers in Antithrombotic Trials

Full access? Get Clinical Tree

Get Clinical Tree app for offline access