177 - The development of cancer immunotherapy and its promise for treating advanced cancers

KEYPOINTS

• 01

  Metastatic solid cancers are the real problem

  Rosenberg says about 600,000 Americans died of cancer in the past year, and roughly 550,000 of those deaths came from solid epithelial cancers (colon, breast, lung, pancreas, prostate and similar). Blood cancers are about 10%. Surgery and radiation cure many localized tumors. Once epithelial cancer has spread, cures are rare, which is why he has spent his career on the immune system.

• 02

  A spontaneous regression set the course of his career

  As a young resident, Rosenberg found a patient whose gastric cancer had spread to his liver and lymph nodes 12 years earlier and been left in place after a palliative operation. The cancer had disappeared without any treatment. Rosenberg concluded that the body had some way to reject cancer, and that the immune system was the likeliest candidate.

• 03

  Chemotherapy has hit diminishing returns for common metastatic cancers

  Chemotherapy came out of mustard-agent research during wartime around 1942. It cures some leukemias, lymphomas and testicular cancer. For most metastatic epithelial cancers, Rosenberg says progress has been incremental. Median survival in metastatic colorectal cancer rose from about 8–10 months to about 2.5 years. Some newer drugs add weeks of life at high cost and toxicity.

• 04

  IL-2 produced the first immune-driven cures

  T-cell growth factor (interleukin-2), discovered in 1976 by Morgan, Ruscetti and Gallo, let researchers keep lymphocytes alive outside the body. Before that, they died within a day or two. Rosenberg's team treated dozens of patients with IL-2 or IL-2-grown cells and saw no responses. After they changed the schedule and raised the doses, a patient with widespread melanoma treated in 1984 had a complete, lasting regression.

• 05

  Only certain cancers responded, and mutations explain why

  High-dose IL-2 caused regressions in about 15–20% of patients with metastatic melanoma and kidney cancer, and some of those were durable cures. Melanoma carries an unusually high mutational burden. Rosenberg's view is that mutated proteins (neoantigens) are what T cells actually recognize.

• 06

  Mutations are the antigens, and they're almost all unique

  In a series of 195 patients with common epithelial cancers, his lab found T cells recognizing mutations in about 80% of them. There were 393 distinct antigens, and almost none were shared between patients. The exception was a KRAS mutation on a rare HLA type. Only about 1.3–2% of a tumor's mutations are immunogenic. He calls this good news: after decades, 'we finally understand what a cancer antigen is.'

• 07

  Tumor-infiltrating lymphocytes improved results, then lymphodepletion jumped them

  Growing TIL cells from tumors in IL-2 and infusing them raised melanoma response rates from about 15% with IL-2 alone to 30–35%, though responses were often short. Wiping out the patient's inhibitory regulatory T cells before infusion pushed response rates to about 55%, with roughly 25% durable complete remissions.

• 08

  CAR T cells were engineered from a gene-insertion idea

  His team inserted receptor genes so T cells could recognize CD19 on B cells. A 2009 patient with chemotherapy-refractory lymphoma, who had a kilogram-sized chest tumor after four failed regimens, was disease-free 12 years later. Losing normal B cells turned out to be survivable.

• 09

  Checkpoint inhibitors release the brakes, but most solid cancers need more

  Anti-CTLA-4 and anti-PD-1 drugs (James Allison later won a Nobel Prize for his checkpoint work) unleash existing immune responses. Rosenberg says most patients with the epithelial cancers that cause 90% of deaths respond poorly, because taking off the brakes does nothing if there aren't enough reactive T cells.

• 10

  The frontier is personalized T cells against each tumor's mutations

  His lab published the first mutation-targeted T-cell treatment in 2014, in a woman with metastatic bile-duct cancer. All of his work now goes into turning these unique reactivities into treatments. He also isolates libraries of T-cell receptors against recurring mutations such as p53 and KRAS, which he calls 'almost an engineering problem.'