Skip to Content

France Elbows Roche Aside on Avastin

The potential off-label use of Avastin for wet age-related macular degeneration has been a source of many arguments over the last few years. The arguments come because another formulation of the same drug, Lucentis, is what’s actually approved for wet AMD, and it’s significantly more expensive. (The two forms appear to have basically the same activity, as determined in a clinical trial that was not looked upon favorably at all by the compounds’ manufacturers and marketers, Roche and Novartis). This has led to a lot of off-label use, and a lot of talk among physicians in various countries about how best to do this.

Now a reader sends word that the government of France has come right out and said it: prescribe Avastin off-label, because Lucentis is equivalent and it’s too costly. This had been tried as a temporary measure earlier this year, to Roche’s displeasure, and now it’s official policy. Lucentis has been a significant expense for the state health service, and the government is striking back with the weapon they have to hand. Expect more of this, especially if companies try to maintain positions like this one. From my perspective, this is a price/efficacy battle that was lost some time ago, and trying to hold on to it just invites this kind of response.


26 comments on “France Elbows Roche Aside on Avastin”

  1. ScientistSailor says:

    Shame on you Derek, I expect much more. Lucentis and Avastin are different molecules, not different formulations of the same molecule. Lucentis is a FAB fragment, that has been affinity matured. Avastin is not manufactured to standards for ocular injection, and is re-packaged for such use in compounding pharmacies. Ask the people who died from spinal fungal infections how well that worked out for them…

  2. SomeDoc says:

    @ScientistSailor Just because Roche/Genentech are willing to make a slightly different product doesn’t make it useful. Nothing was stopping Genentech from seeking an AMD indication for Avastin — or packaging it in smaller dose vials — except the profits from Lucentis.

    We see lots of patients with DME who aren’t covered by medicare/medicaid — and $1500+ per Lucentis injection, repeated monthly for 3-4 months is more money than they make in a year.

  3. flem says:

    Lucentis is not a truly innovative discovery that merits to be reimbursed more than Avastin. I don’t know the full history of Lucentis development but I presume it was started after off-label use of Avastin demonstrated efficacy in the eye. If Roche developed a novel molecule (rather than Avastin itself) they did so hoping to show a clinical advantage. Since that was not accomplished why are we being asked to pay for a molecule no better than a much cheaper alternative? If we are going to solve the raising cost of healthcare let’s make sure we only pay for true innovations that give us value for our HC$ spent. To be clear I do not blame Roche or Novartis for their asking price…(which is what they believe the payer will accept to pay). However I do blame payers who accept to pay an unwarranted price (presuming they do not have a vested interest in the profit generated by a high price- as in the case of the French authorities)

  4. ScientistSailor says:

    @SomeDoc what Derek wrote was factually inaccurate Lucentis and Avastin are totally different molecules.

    @flem Lucentis was in development long BEFORE Avastin was approved. It is a different molecule, and required a completely separated clinical development program.

  5. Hap says:

    If Nexium were sold at twenty times the price of omeprazole for use as a headache remedy, with clinical data for both, then how many people do you think would be buying Nexium for that use? My guess is, close to zero. So, then, why price it there?

  6. Norton I says:

    I worked at the Genentech facility manufacturing Lucentis for a year. Avastin and Lucentis are different molecules that do the same thing, but were developed as different programs, and the manufacturing, formulation, and packaging are very different.

    Official encouragement of off label use, while banning advertisement of off label uses seems like it creates ethical issues.

  7. Not Today says:


    Since Lucentis is a fragment derived from the same antibody as Avastin, saying they are “totally different” is factually inaccurate. Identical, no, but not totally different.

    Since the average drug discovery timeline is 10+ years, saying that “Lucentis was in development long BEFORE Avastin was approved” is not by itself particularly meaningful. What does matter is that Avastin beat Lucentis to market by more than two years.

  8. Magrinho says:

    REGN is eating DNA’s lunch (and dinner and dessert) for this indication anyway.

    Eylea – best drug for AMD but $$$
    Avastin – pretty damn good and cheap
    Lucentis – Meh and $$$

  9. Hap says:

    @6: That seems like more of a concern; while pharmas have more incentive to oversell with off-label use, they are likely in most cases to have better information about such uses, as well, then any of the alternative people (governments, insurance companies) wanting to sell off-label. It is also hypocritical of them (although that doesn’t seem to stop anyone). In this case, though, there should be at least reasonable data to say how they work, side-effect profiles, and so on, compared to one another, so that it isn’t just an ad and a stab in the dark.

    If the market is as @8 says, then Lucentis is hosed as stands. Mediocre and expensive is going to lose badly to cheap and decent or really good and expensive. Sunk costs are sunk.

  10. Drug A says:

    Two points:
    1. If Avastin is to be used off-label for AMD, it has to be repackaged under conditions that don’t put people at risk of horrible eye infections. That’s a surmountable problem, but the solution isn’t to have doctors doing it in their offices.

    2. There is no moral case that drugs should be priced to recoup investment (ok, there is, but it has an upper limit). Drugs should cost what they are worth to patients or society (and patients or society shouldn’t buy them if they aren’t worth it). After years of debate about Avastin v Lucentis, we are no closer to an agreement about what we should be willing to pay for an effective treatment for AMD, but if we accept, for the sake of argument, that Lucentis is over-priced, it does not follow that Avastin’s cost should just be pro-rated from the cost of a dose for colon cancer. Maybe if we had better value-based pricing models, a future Genentech might simply bring the same molecule forward at two different price points.

  11. prezcamacho says:

    Lucentis was developed under the hypothesis that a fab would have better penetrance in the back of the eye resulting in better activity. This was supported by monkey studies. In the context of human AMD that didn’t hold up and Avastin is just as good.

    Eyelea is more effective than either so Lucentis was on a bad glide path in the market anyway.

  12. FJ says:

    @Norton and @ScientistSailor

    Backing up what @NotToday is saying, they’re basically the same thing. The light chains between them are 99.5% identical, differing by one amino acid. The heavy chains (in the fragment region only) are 97.8% identical, differing by five very similar amino acids.

    A review of how Genentech discovered/produces each of these antibodies can be found here (doi:10.1038/eye.2011.66).

    Both were derived from A.4.6.1, an anti-VEGF murine antibody. While both drugs were apparently derived from separate humanized fragments, they likely converged on two very similar structures, as evidenced by their sequences.

    They are indeed made in different conditions, with bevacizumab being produced in CHO cells and ranibizumab being produced in E. coli. However, I don’t know of anything that says that resulting changes in post-translational modifications or what have you translates to significant changes in efficacy when used intraocularly for AMD.

    Also, in terms of safety, the BMJ article that Derek references links to this (DOI: 10.1002/14651858.CD011230.pub2) from the Cochrane Collaboration, which shows pretty convincingly that there’s no real safety difference between using the two for AMD.

  13. 4 ur Is only says:

    The French paper mentioned that the government there is also investigating possible market manipulation by Roche and Novartis around these two products and is considering legal charges.

  14. Kristof says:

    FJ: “bevacizumab being produced in CHO cells and ranibizumab being produced in E. coli. However, I don’t know of anything that says that resulting changes in post-translational modifications” – bevacizumab is made in CHO because it has a glycosylation site on heavy chain, while ranibizumab being FAb fragment doesn’t have one, so it can be made in E.coli.

    1. HT says:

      A bit unfair to cut FJ in mid-sentence. The complete sentence reads “… resulting changes in post-translational modifications or what have you translates to significant changes in efficacy when used intraocularly for AMD.” But your clarification is greatly appreciated.

      Anyway, I don’t see what’s wrong with the current situation, as doctors, payers and patients are not obliged to maximize the pharma’s profits. The obligation lies with the pharma executives, who bet that the payer/ gov’t would cough up more $$$ instead of changing the rules. And they lost the bet in France. Dommage

  15. CMCguy says:

    France is seeming bending rules for drug marketing approvals and probably behaving like a 3rd world country that uses compulsory licensing to get cheaper drugs. As others have pointed out the development, formulation and manufacturing focus and then handling requirements are differently tailored and while these alone may not fully support the cost delta ignoring them is also troubling for promotion of innovation.

  16. Hap says:

    If this be innovation, then don’t let the door hit it in the butt on the way out.

    Pricing power is finite, and expecting to sell nearly identical products (in content and effect) for drastically different prices is asking for your customers to perform arbitrage. Had they priced Lucentis somewhat differently (close to Avastin – say 2 or 3 times as much instead of like twenty), we wouldn’t be discussing this, probably. How many Navigators would Lincoln have sold if they priced them at twenty times the cost of the Explorer (of which it was a luxurified copy), in a market with the Escalade? Would its Edsel-scale failure be a failure to support innovation, or greed and stupidity on Lincoln’s part?

  17. Kristof says:

    @HT – I had nothing to say about PTMs and efficacy therefore I cut the sentence to only show the thing I’m refering to.

    I think glycosylation – or lack of thereof – in monoclonal antibodies has more to do with safety i.e. recognition of antibody as foreign molecule or not.

    By the way, here you can find more information about both molecules:

  18. Kristof says:

    @HT – I had nothing to say about PTMs and efficacy therefore I cut the sentence to only show the thing I’m refering to.

    I think glycosylation – or lack of thereof – in monoclonal antibodies has more to do with safety i.e. recognition of antibody as foreign molecule or not.

  19. tangent says:

    If this be innovation, then don’t let the door hit it in the butt on the way out.

    I laughed, but that seems a *little* hard on the Lucentis development. I’d be willing to award the development work some non-zero points in the Innovation category — just none to the product for Impact.

    Bummer. It happens, and then you don’t get paid. If you want to profit in the capitalist system, there’ll be no playing “market value you pay me for that, no market value you pay me for my labor.”

  20. petros says:

    Following the Italian legal cases last December!

    “an Italian regional court is upholding charges levied against the companies earlier this year, fining Roche and Novartis €180 million ($221.6 million) for manipulating sales of their eye medication.

    The Lazio regional administrative court dismissed appeals from the two companies, which were accused of colluding to hamper use of Avastin to reap the profits off Lucentis, Italian newspaper ANSA reports. The companies’ alleged cartel is estimated to have cost Italy’s national health service over €45 million ($55 million) in 2012 and could have racked up future costs of €600 million a year, the newspaper notes. In March, the Italian Competition Authority imposed penalties of €92 million ($113 million) on Novartis and €90.5 million ($111.4 million) on Roche for their marketing practices.”

  21. Hap says:

    No, there’s innovation there, but innovation isn’t enough, necessarily. Expecting people to pay when the innovation doesn’t work so much (or works no better than a similar product for the same indication) is a bad idea.

    It seems like the same lesson of supplements – no one’s going to go to the trouble of getting real data on a supplement for a specific disease because, while that information would be useful, you can’t be sure that the data would be good enough to make it worth paying for, and other people would be selling it for different indications much cheaper, so that selling it for the appropriate price (because you need to make the money back and because you took the risk to prove your claims) would be impossible.

  22. CMCguy says:

    Hap glad you can at least acknowledge some innovation and I was not actually supporting the cost between the two products (directly at least) but more indicating overlooking a value to incentivize drug R&D. Even if addressing the same VECF-A target Lucentis it appears was specifically tailored for ocular use which based on info involved paring down the molecule and formulation efforts as successful improvements for treatment of AMD. Neither of these actions may have taken great leaps of creativity to accomplish based on info at hand however a direction was proposed then done followed thorough clinical testing performed and approval obtained. The manufacturing requirements are different by design hence not really like a generic vs brand drug. Why would anyone go down that road, especially for small populations? Its easy to retroactively suggest all this extra effort was unnecessary because another drug approved for other indications apparently works almost as well, especially if you ignore re-compounding risks. Although France does have and uses power to set pricing what they did here was basically grant a defacto approval of Avastin for a non approved label claim where industry gets pounded on for similar extensions without running full-fledged trials.

  23. Hap says:

    They couldn’t have known it would turn out that way ahead, and they took a risk in doing so. The risk, though, doesn’t mean they have to get paid – if the risk doesn’t pan out, then they aren’t obligated to demand people pay for it. (Alternatively, had they priced it differently, they could still have gotten paid for that risk – but they demanded the sky and stars for something not reasonably worth them.)

  24. Hap says:

    Depending on how much leverage they have with Avastin, I imagine that if they feel they’ve priced Lucentis appropriately, they could simply raise the price of Avastin to inhibit buying it as a Lucentis replacement. If you judge the innovation to be in Avastin, well, then they would be pricing it accordingly. I don’t think that would make people any happier, and would force the pricing issue to the fore, but that’s how this is all going anyway. Alternatively, they could either lower the price of Lucentis (to 2 or 3 times Avastin) or try to approve an Avastin opthalmic dose and charge similarly.

    Innovation isn’t worth anything on its own – it has to do something for others, and do something for them proportionate to what you charge/they pay for it.

Comments are closed.