#262: Could Gene Therapy Increase the Risk of Cancer, & More
1. Could Gene Therapy Increase the Risk of Cancer?
This week, Blue Bird Bio (BLUE) announced that it put on hold two of its sickle cell gene therapy trials after two patients developed different blood cancers: acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS).
Scientists have offered several theories for the causes of these cancers, among them the following:
1. Lentiviral Vectors (LVV): LVVs delivered Blue Bird Bio’s gene therapy for sickle cell disease and could have activated an oncogene, a gene that can transform into a cancer cell, or could have deactivated a gene that suppresses tumors. This week, Blue Bird announced that, according to its postmortem, LVVs did not likely cause the AML.
2. Cancer Predisposition: compared to the general population, sickle cell patients are 72% more likely to develop hematological malignancies. The risk to female patients and adult patients aged 15-39 years is three-fold that of other sickle cell disease patients.
3. Hydroxyurea: physicians often prescribe hydroxyurea to sickle cell patients to prevent red blood cells from transforming into a sickle shape. At times, hydroxyurea can be toxic.
4. Busulfan: as it preconditions patients undergoing bone marrow transplants to accept newly transplanted cells, busulfan can be quite toxic. Based on a recent study from the Center for International Blood and Marrow Transplant Research, the cumulative incidence of cancers after hematopoietic stem cell transplantations was 0.6% and 1.2%, respectively, after 5 and 10 years.
While all four of these explanations may have played a role in causing cancer, scientists are trying to control one potential cause by developing less toxic pre-conditioning regimens. Today, busulfan is an inexpensive generic drug. We believe that if scientists can create a potent but less toxic pre-conditioning regimen, the higher sticker price it will command should be well worth it.
2. Tech Giants Are Fighting a Proxy War Over Privacy Standards

In June 2020, Apple announced that it will ask users to opt in if they would like to share their ID for Advertisers (IDFA). IDFA is a unique identifier in Apple mobile devices enabling advertisers to track users across the web and to target them with personalized ads. Having planned to make the change effective with the launch of iOS 14 in September 2020, Apple faced a massive backlash and delayed it until early 2021. As the deadline looms, here are what we believe to be the arguments for and against IDFA from the view of various constituencies.
‘Facebook & Small Businesses’
If Apple’s 1 billion mobile users opt-out of sharing their IDFAs, small, medium, and large businesses will not be able to serve personalized ads. As Facebook claims, “Yes, there will be an impact to Facebook’s diversified ads business, but it will be much less than what will befall small businesses.” Unable to serve personalized ads, small businesses will be forced to spend more on advertising per sales dollar. Based on a Facebook study, the shift to IDFA could hit 60% of small business sales.
‘Apple & The Consumer’
Apple believes its privacy-first standards will continue to benefit consumers. At Apple’s Data Privacy Day in January, Craig Federighi, senior VP of Software Engineering, stated, “Privacy means peace of mind, it means security, and it means you are in the driver’s seat when it comes to your own data.”
‘Reality?’
Facebook is advocating for small businesses because they dominate its revenue base. Apple is fighting for privacy standards that impact competitors’ advertising platforms but do not impact its personalized ad platform. Both Apple and Facebook are fighting for what they believe is right, but both have ulterior motives which we believe are important to highlight.
3. Starlink Is Focusing on the Mobility Market

This week, SpaceX sought FCC approval to connect “vehicles, vessels and aircraft” to its Starlink constellation. SpaceX noted that the antennae would be identical to those in its consumer terminals but would have different mountings. Musk explained further that the antennae would not connect Tesla cars to Starlink but would target aircraft, ships, and large trucks/RVs.
One of the biggest hurdles to the proliferation of low earth orbit broadband is the cost of user terminals. While rocket launch and satellite costs have dropped dramatically, terminals remain prohibitively expensive. According to ARK’s research, current phased array antennae cost $7,000-20,000, preventing widespread household adoption.
According to Wright’s Law, scaling unit production is key to lower costs. In the early days of innovation, use cases at high price points are critical.
Accordingly, SpaceX’s Starlink has targeted the mobility market. Based on ARK’s research, airplane antennae currently cost hundreds of thousands of dollars while those for boats and trucks are in the tens of thousands of dollars. ARK estimates that the market for connecting planes, trains, and motor vehicles will be $36 billion by 2025.
4. Coupang Is at the Forefront of Delivering a Frictionless E-commerce Experience
This week Coupang, a Softbank-backed Korean e-commerce leader, debuted on the New York Stock Exchange (NYSE). Like Amazon and JD.com, Coupang has built an integrated supply-chain backbone that enables frictionless e-commerce fulfillment. Coupang delivers 99%+ of its orders within 24 hours, the average delivery taking several hours, trouncing the value proposition of convenience stores. Users can return merchandise by placing reusable bags outside their apartment doors, receiving refunds at the time of pickup. Founded by Bom Kim and enabled not only by the dense population but also by the dearth of retail infrastructure in South Korea, Coupang deserves credit for delivering what we believe is one of the best ecommerce user experiences in the world.
5. Autonomous Systems Should Be Judged by Functionality

The Society of Automotive Engineers has defined five Levels of autonomous driving, 5 being the highest. While many analysts evaluate autonomous driving solutions based on these somewhat arbitrary levels, ARK believes that functionality is more relevant.
Recently, Honda announced that it has developed the world’s first Level 3 autonomous car and that it intends to produce 100 Legend vehicles with Sensing Elite technology in Japan. Meanwhile, in a shift from its cautious stance on autonomous driving, the Japanese government now will allow drivers to take their eyes off the road when using the Traffic Jam Pilot function. Traffic Jam Pilot works when a vehicle is “caught in traffic congestion, under certain slow speed conditions.”
Including its Full Self Driving software update, Tesla’s Autopilot is currently defined as a Level 2 autonomous system, requiring the driver to maintain full control of the car with his or her eyes on the road and hands on the wheel. Our research indicates autopilot already has more functionality than the Honda “Level 3” system, including Summon, Autopark, and Traffic and Stop Sign Control beta features. ARK believes that an autonomous vehicle’s functionality is a better marker of progress than are SAE Levels.