Biotech Values

[DewDiligence]

How will Xarelto/Eliquis/Pradaxa affect US Lovenox sales?

[Updated for phase-3 Eliquis results in secondary VTE prevention.]


To answer this question, let’s divide the US Lovenox market into its various indications and then assess the proportion of Lovenox sales in each indication that is at-risk from competition by the new oral anticoagulants.

(In this analysis, I make no distinction between SNY’s branded Lovenox and NVS/MNTA’s generic Lovenox. Although competition from the new oral anticoagulants may initially take proportionally more share from the generic, I expect the SNY-NVS Lovenox duopoly to re-establish itself as close to a 50/50 split in overall sales once the effect of competition from the new oral drugs becomes clear.)

SNY has disclosed that approximately 60% of US Lovenox sales (by dollars, not number of Rx’s) come from hospital settings and 40% come from outpatient settings. Based on this information, I estimate the following breakdown of US Lovenox sales by indication:

 
                                   Proportion of  
                                ---Lovenox Sales---  
 Indication*                    Hospital Outpatient 
  
 A) Primary VTE prevention  
 following hip/knee surgery       10%       30%  
  
 B) Primary VTE prevention  
 for immobilized patients         30%  
  
 C) Secondary VTE prevention  
 (prevention of recurrence)                 10%  
  
 D) Acute VTE treatment           10%  
  
 E) ACS                           10%  
  
 =============================    ===       ===  
 All indications                  60%       40%  
  
 *Lovenox is not FDA-approved for stroke prevention in 
 atrial fib, where the new oral anticoagulants will likely  
 find their largest market as replacements for warfarin. 
 Although Lovenox gets some off-label use as a “bridging” 
 agent to warfarin in this indication, my model excludes  
 such off-label use for the sake of simplicity.

Now let’s take these five indications one at a time and estimate how much share the new oral anticoagulants will take from Lovenox when they become established.

A) Primary VTE prevention following hip/knee surgery. This is a strong indication for the new oral anticoagulants because the risk of thrombosis immediately following implant surgery is sufficiently high that the bleeding risk of the new oral anticoagulants is easy to justify. Xarelto was just approved by the FDA in this indication: #msg-64830766. (Pradaxa is approved in this indication the EU [#msg-27956748] but not the US, and the status of Eliquis is unclear [#msg-58061618].) As a once-daily pill, Xarelto should be able to take a considerable share of the (post-discharge) outpatient sales of Lovenox in this setting; however, Lovenox’s (pre-discharge) hospital sales in this setting should hold up better due to Lovenox’s longstanding safety record, its lower copayment costs (due to its availability as a generic :- )), and its injected formulation, which is preferable to a pill in the hospital setting. Based on feedback from urche (#msg-64860129), I estimate that Xarelto will take 60% of Lovenox’s outpatient sales and 50% of its hospital sales in this setting; according to the numbers in the table above, this yields (0.6)(0.30) + (0.5)(0.10) = 0.23 — i.e. a 23% loss of overall US Lovenox sales attributable to primary VTE prevention following hip/knee surgery.

B) Primary VTE prevention for immobilized patients. Unlike VTE prevention following hip/knee surgery (A), where the risk of thrombosis per unit of time is high in the period immediately following surgery, the risk of thrombosis per unit of time in non-surgical immobilized patients is relatively low, which makes the bleeding risk of the new oral anticoagulants harder to justify in this setting. Xarelto failed in this indication in a phase-3 trial called MAGELLAN (#msg-61803303) and Eliquis failed in a phase-3 trial called phase-3 trial called ADOPT (#msg-68938689). Pradaxa has not (to my knowledge) been tested in this indication. Thus, I expect little if any loss of Lovenox sales in this hospital-based indication; for this exercise, I assume zero loss of Lovenox sales in primary VTE prevention for immobilized patients.

C) Secondary VTE prevention. I previously though the new oral anticoagulants are badly suited to this indication for essentially the same reason as in B): the risk of thrombosis per unit of time is too low to justify the increased risk of serious bleeding; indeed this proved to be the case with Xarelto (#msg-44245112). However, Eliquis generated stellar data in secondary VTE prevention at ASH 2012 (#msg-82279247), and I now think it could usurp half of Lovenox’s US business in this indication—i.e. a 5% loss of overall US Lovenox sales attributable to secondary VTE prevention.

D) Acute VTE treatment. The new oral anticoagulants are well suited to this indication for essentially the same reason as in A): the risk of thrombosis per unit of time is sufficiently high that the bleeding risk of the new oral anticoagulants is easy to justify. Although Xarelto has a CHMP approval for this indication (#msg-67357050), JNJ (Bayer’s US partner) has not yet submitted an NDA to the FDA, nor has JNJ stated an intention to do so. Eliquis is being tested in two phase-3 trials due to finish in late 2012 or early 2013 (http://clinicaltrials.gov/ct2/show/NCT00643201 , http://clinicaltrials.gov/ct2/show/NCT00633893 , #msg-29928836). Pradaxa has completed two phase-3 trials: the first reported data in 2009 (#msg-4244839) and the second should report data soon (http://clinicaltrials.gov/ct2/show/NCT00680186 ). The comparator in both Pradaxa trials is warfarin rather than Lovenox; however, uptake of Pradaxa in this indication at the expense of warfarin could cut into Lovenox’s use as a “bridging” agent to warfarin. All told, I expect the new oral drugs to take about 50% of Lovenox’s sales in this indication, which means a (0.5)(0.10) = 5% loss of overall US Lovenox sales attributable to acute DVT treatment.

E) ACS. ACS, which is a catchall term for STEMI, NSTEMI, and unstable angina, includes an acute segment and a chronic segment; Lovenox is commonly used in the former, but not the latter. The only one of the new oral anticoagulants that achieved nominal success in ACS is Xarelto, but the FDA rejected the NDA in Jun 2012 (#msg-76856261). Moreover, the phase-3 Xarelto trial in ACS was limited to secondary prevention and hence is of little consequence to Lovenox (#msg-69172520). Eliquis failed in phase-2 due to excess bleeding without a concomitant decrease in ischemic events (#msg-56880416). Pradaxa also failed in phase-2 due to excess bleeding without a concomitant decrease in ischemic events (http://clinicaltrials.gov/ct2/show/results/NCT00621855 ); a separate small Pradaxa phase-2 trial in PCI finished in 2010 (http://clinicaltrials.gov/ct2/show/NCT00818753 ), with no announcement of plans to advance to phase-3, so it’s reasonable to presume that there aren’t any. All told, I assume zero loss of overall US Lovenox sales to the new oral anticoagulants in ACS.

--
Summing the five cases above yields a 33% overall loss of US Lovenox sales to market share gains by the new oral anticoagulants once these new drugs are established. (23% of this 33% figure comes from VTE prevention following hip/knee surgery, 5% comes from secondary VTE prevention, and 5% comes from acute VTE treatment.) Given a typical sales ramp for expensive new drugs, it will likely take several years until the new oral anticoagulants are able to reach their potential in the indications discussed above, so the impact of these drugs on US sales of Lovenox will presumably be gradual.