No­var­tis, UCSF re­searchers urge FDA to clean up its messy ad­verse events data­base

A team of re­searchers from No­var­tis, Or­a­cle Health Sci­ences and the Uni­ver­si­ty of Cal­i­for­nia, San Fran­cis­co say the FDA’s ad­verse drug re­ac­tion data­base could be im­proved by group­ing drugs by their chem­i­cal struc­ture and au­tomat­ing cer­tain re­port­ing func­tions.

In a pa­per ap­pear­ing in eLife ear­li­er this month (ti­tled: Re­verse trans­la­tion of ad­verse event re­ports paves the way for de-risk­ing pre­clin­i­cal off-tar­gets) the re­searchers say such changes could help ad­dress sev­er­al ma­jor chal­lenges to in­ter­pret­ing da­ta in FDA’s ad­verse event re­port­ing sys­tem known as FAERS.

Specif­i­cal­ly, the au­thors say that the abil­i­ty to au­to­mat­i­cal­ly map drugs by in­gre­di­ent could make it eas­i­er to iden­ti­fy trends and sift through some of the noise in­her­ent in the open-sub­mis­sion data­base. The au­thors al­so say that in­tro­duc­ing au­toma­tion to the re­ports them­selves could cut down on the num­ber of er­ro­neous or mis­clas­si­fied re­ports.

FAERS cur­rent­ly con­tains well over 8.5 mil­lion re­port­ed ad­verse drug re­ac­tions and is grow­ing at an in­creas­ing rate, with the num­ber of ADRs re­port­ed to FDA ris­ing from less than 100,000 per quar­ter in the late-1990s through the mid-2000s to around 300,000 per quar­ter in 2015.

The ma­jor­i­ty of re­ports to FAERS orig­i­nate from pa­tients (40%) and health­care providers (physi­cians, 25%; oth­er health pro­fes­sion­al, 16%; and phar­ma­cist 5%); the re­main­ing re­ports come from lawyers (3%) or did not iden­ti­fy the re­porter (9%).

Ac­cord­ing to the au­thors, those re­ports match up to just over 7 mil­lion in­di­vid­ual ADRs af­ter ac­count­ing for mul­ti­ple re­ports about the same in­ci­dent and du­pli­cate re­ports.

Too Many Names

One of the biggest fac­tors that com­pli­cates re­search in­to FAERS da­ta is the fact that many drug in­gre­di­ents are sold un­der mul­ti­ple names and in dif­fer­ent for­mu­la­tions.

To get around this, the au­thors ag­gre­gat­ed drugs by their ac­tive in­gre­di­ent, which they say pro­vid­ed stronger drug-ADR sig­nals than by ag­gre­gat­ing based on syn­onym group­ing alone.

This pro­vid­ed the au­thors with a list of 2,729 unique ac­tive in­gre­di­ents. Of those, the au­thors found that a sur­pris­ing 30% had no re­port­ed ADRs and that 90% of the re­port­ed ADRs were at­trib­uted to just un­der half the in­gre­di­ents found in FAERS.

In the case of flu­ox­e­tine, the ac­tive in­gre­di­ent in Prozac, the au­thors found near­ly 400 “syn­onyms” for the drug. Over­all, the au­thors found an av­er­age of 16 syn­onyms per ac­tive in­gre­di­ent in FAERS.

With­out a big­ger pic­ture view of ADRs for drugs with nu­mer­ous syn­onyms, the au­thors say that cer­tain side ef­fects may false­ly ap­pear to be sig­nif­i­cant or in­signif­i­cant. For ex­am­ple, sex­u­al dys­func­tion, a well-known side ef­fect of flu­ox­e­tine, ap­peared to be sta­tis­ti­cal­ly in­signif­i­cant when look­ing on­ly at re­ports for Prozac.

But when look­ing across all ag­gre­gat­ed da­ta for flu­ox­e­tine, the au­thors found that sex­u­al dys­func­tion stood out clear­ly.

Re­port­ing Bias

The au­thors al­so found that re­ports in FAERS dis­pro­por­tion­ate­ly tilt­ed to­wards se­ri­ous or life-threat­en­ing out­comes.

Near­ly 13% of ADRs they iden­ti­fied were to re­port a death, 5% re­port­ed a life-threat­en­ing event, and 34% re­port­ed a hos­pi­tal­iza­tion. But 42% of re­port­ed out­comes fell un­der “oth­er,” which the au­thors say is the on­ly one of FAERS sev­en out­come op­tions that is fit for re­port­ing non-se­ri­ous out­comes.

“It is a fea­ture of re­port­ing in an open sub­mis­sion data­base like FAERS that this ra­tio does not re­flect the true bal­ance be­tween fa­tal and rel­a­tive­ly be­nign drug ADRs, but rather the ra­tio of the ADRs that are thought to mer­it re­port­ing,” the au­thors write.


In many cas­es, the au­thors say that re­porters con­fused ad­verse re­ac­tions with a drug’s in­di­ca­tion.

“Ap­prox­i­mate­ly 5% of all re­ports for any drug de­scribe the drug’s in­di­ca­tion as an ad­verse event,” the au­thors write.

For in­stance, some re­ports iden­ti­fied di­a­betes as a side ef­fect for the drug rosigli­ta­zone, which is used to treat type two di­a­betes.

The au­thors sug­gest that re­porter ed­u­ca­tion could play a role in re­duc­ing the num­ber of re­ports that con­fuse in­di­ca­tion and side ef­fect, as the num­ber of such re­ports be­gan to drop in 2011 when FDA’s 2010 fi­nal rule on ad­verse event re­port­ing went in­to ef­fect.

The au­thors al­so found that oth­er fac­tors, in­clud­ing reg­u­la­to­ry ac­tiv­i­ty and news re­ports can in­flu­ence ad­verse event re­port­ing in ways that can make in­ter­pret­ing trends dif­fi­cult.

In the case of rosigli­ta­zone and an­oth­er re­lat­ed drug, pi­ogli­ta­zone, the au­thors ob­served a sta­tis­ti­cal­ly sig­nif­i­cant sig­nal be­tween both drugs and car­diac events.

But while the au­thors found that the num­ber of heart-re­lat­ed re­ports for rosigli­ta­zone was fair­ly con­stant over time, such re­ports for pi­ogli­ta­zone peaked along­side a pe­ri­od of in­creased scruti­ny of rosigli­ta­zone and re­mained low oth­er­wise.

First pub­lished here. Reg­u­la­to­ry Fo­cus is the flag­ship on­line pub­li­ca­tion of the Reg­u­la­to­ry Af­fairs Pro­fes­sion­als So­ci­ety (RAPS), the largest glob­al or­ga­ni­za­tion of and for those in­volved with the reg­u­la­tion of health­care and re­lat­ed prod­ucts, in­clud­ing med­ical de­vices, phar­ma­ceu­ti­cals, bi­o­log­ics and nu­tri­tion­al prod­ucts. Email for more in­for­ma­tion.

Mov­ing Out of the Clin­ic with Dig­i­tal Tools: Mo­bile Spirom­e­try Dur­ing COVID-19 & Be­yond

An important technology in assessing lung function, spirometry offers crucial data for the diagnosis and monitoring of pulmonary system diseases, as well as the ongoing measurement of treatment efficacy. But trends in the healthcare industry and new challenges introduced by the COVID-19 pandemic are causing professionals in clinical practice and research to reevaluate spirometry’s deployment methods and best practices.

Paul Hudson (Getty Images)

Sanofi, Glax­o­SmithK­line jump back in­to the PhI­II race for a Covid vac­cine — as the win­ners con­gre­gate be­hind the fin­ish line

Sanofi got out early in the race to develop a vaccine using more of a traditional approach, then derailed late last year as their candidate failed to work in older people. Now, after likely missing the bus for the bulk of the world’s affluent nations, they’re back from that embarrassing collapse with a second attempt using GSK’s adjuvant that may get them back on track — with a potential Q4 launch that the rest of the world will be paying close attention to.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 105,400+ biopharma pros reading Endpoints daily — and it's free.

SCO­TUS de­clines to re­view En­brel biosim­i­lar case, tee­ing up 30+ years of ex­clu­siv­i­ty and $20B more for Am­gen’s block­buster

As the House Oversight Committee is set to grill AbbVie CEO Richard Gonzalez on Tuesday over tactics to block competition for its best-selling drug of all time, another decision on Capitol Hill on Monday opened the door for billions more in Amgen profits over the next eight years.

The Supreme Court on Monday denied Novartis subsidiary Sandoz’s petition to review a Federal Circuit’s July 2020 decision concerning its biosimilar Erelzi (etanercept-szzs), which FDA approved in 2016 as a biosimilar to Amgen’s Enbrel (etanercept). Samsung’s Enbrel biosimilar Eticovo also won approval in 2019 and remains sidelined.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 105,400+ biopharma pros reading Endpoints daily — and it's free.

No­var­tis' En­tresto takes its 2nd fail­ure of the week­end at ACC, show­ing no ben­e­fit in most dire heart fail­ure pa­tients

Novartis’ Entresto started the ACC weekend off rough with a trial flop in heart attack patients, slowing the drug’s push into earlier patients. Now, an NIH-sponsored study is casting doubt on Entresto’s use in the most severe heart failure patients, another black mark on the increasingly controversial drug’s record.

Entresto, a combination of sacubitril and valsartan, could not beat out valsartan alone in an outcomes head-to-head for severe heart failure patients with a reduced ejection fraction (HFrEF), according to data presented Monday at the virtual American College of Cardiology meeting.

How to man­u­fac­ture Covid-19 vac­cines with­out the help of J&J, Pfiz­er or Mod­er­na? Bi­ol­yse sees the dif­fi­cul­ties up close

When Biolyse, an Ontario-based manufacturer of sterile injectables, forged a deal with Bolivia last week to manufacture up to 50 million J&J Covid-19 vaccine doses, the agreement kicked off what will prove to be a test case for how difficult the system of compulsory licenses is to navigate.

The first problem: When Biolyse asked J&J, via a March letter, to license its Covid-19 vaccine, manufacture it in Canada and pay 5% royalties on shipments to needy, low-income countries, J&J rejected the offer, refusing to negotiate. J&J also did not respond to a request for comment.

Tim Mayleben (L) and Sheldon Koenig (Esperion)

On the heels of a sting­ing Q1 set­back, Es­pe­ri­on's long­time cham­pi­on is ex­it­ing the helm and turn­ing the wheel over to a mar­ket­ing pro

Just days after getting stung by criticism from a badly disappointed group of analysts, there’s a big change happening today at the helm of Esperion $ESPR.

Longtime CEO Tim Mayleben, who championed the company for 9 years from early clinical through a lengthy late-stage drive to successfully get their cholesterol drug approved for a significant niche of patients in the US, is out of the C suite, effective immediately. Sheldon Koenig — hired at the end of 2020 with a resume replete with Big Pharma CV sales experience —  is stepping into his place, promising to right a badly listing commercial ship that’s been battered by market forces.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 105,400+ biopharma pros reading Endpoints daily — and it's free.

Matt Gline (L) and Vivek Ramaswamy

In­sid­er ac­count of Roivan­t's SPAC deal — and that $7.3B val­u­a­tion — re­veals a few se­crets as Matt Gline po­si­tions the com­pa­ny as the new ‘Big Phar­ma’

It was Oct. 7, 2020, and Matt Gline wasn’t wasting any time.

The CEO of Roivant had word that KKR vet Jim Momtazee’s SPAC had priced late the night before, triggering a green light for anyone interested in pursuing a big check for future operations and riding the financial instrument to Nasdaq. So he wrote a quick email congratulating Momtazee, whom he knew, for the launch.

Oh, and maybe Momtazee would like to schedule something with Gline and his executive chairman, Roivant founder Vivek Ramaswamy, to chat about Roivant and its business?

Endpoints Premium

Premium subscription required

Unlock this article along with other benefits by subscribing to one of our paid plans.

Re­gen­eron's Evkeeza shows promise in curb­ing high triglyc­erides, but will ge­net­ic dis­par­i­ties lim­it use?

When Regeneron scored an early approval for lipid lowering antibody Evkeeza back in February, the drugmaker cracked open a new pathway to lower abnormally high cholesterol levels. Now, Regeneron is chasing high triglycerides as well with some promising mid-stage data — but will genetic restrictions limit the drug’s use?

Regeneron’s Evkeeza (evinacumab) cut median triglyceride levels by more than 800 mg/dL (57%) in patients with a rare disorder causing abnormally high triglyceride levels compared with an overall increase of 50 mg/dL (1.8%) in participants on placebo, according to Phase II data presented Sunday at the virtual American College of Cardiology meeting.

Pascal Soriot, AstraZeneca CEO (AP Images)

As­traZeneca's Farx­i­ga missed big on Covid-19 study, but it's tak­ing SGLT2 safe­ty da­ta as a sil­ver lin­ing

AstraZeneca hasn’t seen many setbacks in recent months for SGLT2 inhibitor Farxiga, which broke ground in heart failure and kidney disease regardless of diabetes diagnosis. But the British drugmaker had to admit defeat in taking Farxiga into Covid-19. However, follow-up results add a bit of a silver lining to that trial’s safety data.

Of hospitalized Covid-19 patients dosed with AstraZeneca’s Farxiga, 11.2% experienced an organ failure or died after 30 days of therapy compared with 13.8% of those given placebo, according to follow-up data from the DARE-19 study revealed Sunday at the virtual American College of Cardiology meeting.