When One Drug Is Not Enough: How a Four-Drug Prescription Changed Childhood Leukemia. Frei, Freireich and VAMP (1955-1972)
In this article
When One Drug Is Not Enough: How a Four-Drug Prescription Changed Childhood Leukemia. Frei, Freireich and VAMP (1955-1972)
Yesterday we saw that Einhorn did not leave cisplatin alone but joined it to vinblastine and bleomycin (Day 141). The root of that idea lies not in platinum but in a children's ward. On Day 122 we followed Farber's aminopterin: short remissions that ended in resistance. Today we follow the answer the National Cancer Institute gave to the same question: if one drug is not enough, can several drugs that strike by different routes be given at once? Yesterday's box said "the mid-1960s"; the sources place the start of this work between 1961 and 1964, and 1965 is the year the two physicians left the institute. I open the conflict below.
A disease like malaria
Emil "Tom" Frei was 31 and Emil "Jay" Freireich was 28; both joined the NCI clinical program in 1955 (Cancer History Project). According to Benz and Nathan's tribute to Frei, Gordon Zubrod hired the two and urged them to combine drugs as they became available; Zubrod compared childhood leukemia to malaria. The same piece says that before effective chemotherapy about half of children with acute leukemia lost their lives within three months. Farber's single-agent aminopterin and amethopterin had produced some remissions, but most were short because of resistance (Day 122).
Comparative studies came first. In December 1958 in Blood (volume 13, pp. 1126-1148), Frei, Holland, Schneiderman, Pinkel, Freireich and colleagues randomly assigned 65 patients with acute leukemia to two antimetabolite regimens; giving methotrexate daily or every third day made no difference to the frequency of remission. The authors concluded that comparative treatment trials in acute leukemia could be run without apparent harm to patients. Seventeen patients lost their lives while the study was under way. In October 1961, in the same journal (volume 18, pp. 431-454), 6-mercaptopurine and methotrexate were compared: in children with acute lymphocytic leukemia the remission rate was 59% with the two together, 47% with 6-mercaptopurine alone and 29% with methotrexate alone. The median remission lasted 4-5 months in all three groups and median survival was 9 months.
It should be read like this: the two-drug combination roughly doubled the remission rate compared with methotrexate alone and raised it by 12 points compared with 6-mercaptopurine alone, but it did not change the horizon. The authors note that the combined result was what independent action of the two drugs would predict. The two drugs also had similar toxicity (bone marrow and gut). Something new was needed.
A drug that failed the ruler
According to Freireich's 2001 interview for MD Anderson, the drug arrived through Irv Johnson of Eli Lilly: the company was sifting alkaloids of the periwinkle (Vinca rosea). Vincristine was active against mouse leukemia P388 but not L1210. Freireich says he took it to Zubrod around 1960 or 1961, that Zubrod, by the L1210 yardstick, judged it a total bust, and that he himself, arguing that the children had no other option, obtained permission to work on his own personal recognizance. He adds that in the first 11 children, 7 achieved complete remission on vincristine alone. These are memories from decades later; I could not reach the primary report.
According to Frei's 1972 presidential address, vincristine plus prednisone produced remission in about 90% of childhood acute lymphocytic leukemia. There is also independent confirmation: in the series reported by Hardisty, McElwain and Darby in the BMJ (1969), all 22 children with acute lymphoblastic leukemia given it as first treatment achieved complete remission. Remission was not cure, but the leap was real.
Four drugs at once
According to Freireich the idea was simple: four drugs (vincristine, amethopterin, that is methotrexate, mercaptopurine and prednisone) that broke cells down by different routes and left different side effects could, if given together, add their effects. Colleagues thought it reckless; he says they called him crazy, and many believed cancer could be treated only with surgery and radiotherapy. Zubrod was upset and wanted a randomized trial first; approval came through the review process, with Zubrod and Mider taking part. Freireich sat with the families every Tuesday and Thursday to explain. By his account, of 17 consecutive untreated children, 16 responded within two to three weeks.
Limit: this is one side's account, given decades after the events. The figures differ in other sources (see the fact-check box below).
Why four drugs? Different side effects, full dose
The inference in Frei's 1972 address was this: drugs whose dose-limiting side effects differ can be given together at full dose; drugs whose side effects overlap (methotrexate and 6-mercaptopurine, for example) must have their doses cut when combined, and the effect weakens. VAMP rested on this logic. The same address says VAMP was started by Freireich and colleagues in 1963 as five 10-day courses, that the outcome measure was the length of remission without maintenance treatment, and that VAMP gave significantly longer remissions than single agents. One of the three advances listed in Frei's address was the "exponential iceberg" model: because tumor cell numbers fall exponentially, when remission appears the cells have not all gone. Skipper's work, who reported by 1964 that combination chemotherapy could cure mouse leukemia L1210, fits the same line (ASCO Post, 2024).
Remission is not cure, the price was heavy, and none of it worked without supportive care
Relative and absolute effect should be read together: in 1955 fewer than 1% of children with leukemia lived long (Cancer History Project); those who wrote Freireich's farewell note state that five-year survival in children with acute lymphoblastic leukemia now exceeds 90% (ASCO Post, 2021). The leap was not VAMP's work alone: Frei's farewell note says that adriamycin and asparaginase, added later, brought the cure rate close to 90% (JCI, 2013). By 1970 combination chemotherapy was accepted as curing childhood leukemia, helped by platelet transfusion and clean-air rooms (ASCO Post, 2024). Even in 1975 Freireich says scientists were still debating whether chemotherapy cured leukemia (Cancer History Project).
The price: children were brought to the edge of life by the drugs, and the marrow recovered a few weeks later; because none of the drugs crossed the blood-brain barrier, the disease often relapsed in the nervous system, and that relapse cost the lives of all but about 5% of patients (Wikipedia only; for orientation only). There was opposition too: the Acute Leukemia Group B (ALGB) at first vetoed wide use of VAMP-type protocols as too risky, and Farber spoke against combining cancer drugs (ASCO Post, 2024); Bodey recalls that at first the approach was considered contrary to ethics (MD Anderson narrated video). Half the success lay in supportive care: Freireich says platelet transfusion essentially removed bleeding as a cause of lost lives; that is his own assessment.
Limit of evidence: the first VAMP series was small and not randomized; according to ASCO Post (2024) it was reported in 1964 in an AACR abstract with 25 patients and a 60% complete remission rate. What Zubrod wanted first was randomization. Which patients were durably cured became clear only with long follow-up.
What remained
- The principle of using drugs with different side-effect profiles together at full dose
- Not stopping at remission but continuing treatment; the idea of residual cells
- Comparative, sometimes placebo-controlled trials (in Freireich's account, the 6-mercaptopurine versus placebo trial)
What fell or was limited
- The expectation that one good drug is enough: resistance and short remissions
- The assumption that remission is cure: hidden cells and nervous-system relapse
- The expectation that a mouse model shows every drug in advance: in Freireich's account vincristine failed the L1210 yardstick and worked in humans
Frei and Freireich's road to combination therapy: 1955 to 1972. The axis runs from 1955 to 1975. Green markers are dates found in several sources or in bibliographically verified records (December 1958, October 1961, 1965, December 1972); the amber bar is the interval on which sources conflict (the start of VAMP, 1961-1964); blue markers rest on a single secondary source, on a memoir or on the year alone (1955, the vincristine period, the 1964 abstract). Year-only dates are drawn mid-year and month-only dates mid-month.
A series from two university hospitals
My sources show no direct link between Turkey and this 1960s work, and I do not invent one. There is a measurable present-day data point: the retrospective series of Güneş and colleagues in Annals of Hematology (2014;93(10):1677-1684) followed 343 children with newly diagnosed acute lymphoblastic leukemia treated at two university hospitals in Turkey in 1995-2009 with the ALL-BFM 95 protocol. Complete remission was 97%, five-year event-free survival 78.4%, five-year overall survival 79.9%; relapse 14.8%; 69 of the 343 children (20.1%) lost their lives, mostly to infection and relapse; the abstract states that none were lost to drug toxicity. The inference is mine: the multi-drug protocols in this series descend from the idea described above; the paper does not say so. Two centers and one protocol cannot be carried over to the country as a whole, nor compared directly with the 1% of 1955.
From 1955 to today
What is verified, what conflicts
- Reading limit: I could not read the first primary reports on VAMP: the 1964 AACR abstract only through ASCO Post's account, and I could not reach the vincristine paper by Karon and colleagues at all (its citation could not be verified). For the 1958 and 1961 Blood papers only the abstracts were read. Most of the story is secondary or memoir: the Freireich interview (2001), Frei's 1972 address, JCI and ASCO Post pieces, the MD Anderson narrated video, the Cancer History Project.
- Start of VAMP: Wikipedia says 1961, the MD Anderson narrated video 1962, Frei's 1972 address 1963, ASCO Post shows an AACR abstract in 1964; the Cancer History Project speaks of a "1962 study". Yesterday's tomorrow-box said "the mid-1960s", which was loose: 1965 is only the year the two physicians left the NCI. Amber interval in the figure.
- Patient numbers and outcome: Freireich says 16 of 17 children responded; ASCO Post gives 25 patients and a 60% complete remission rate; in the MD Anderson video Bodey says "eleven" of sixteen patients were cured. These may be different measures, periods or records; I did not merge them.
- First vincristine results: "7 of 11 children" appears only in Freireich's 2001 memory; the publication citation could not be verified.
- About 90% remission: Frei says "about 90%" in 1972; in Hardisty and colleagues' series 22 of 22 children treated first-line. Two separate sources, different patient groups.
- Durable cure rate: ASCO Post (farewell note) "five-year survival above 90%"; Cancer History Project "above 75%"; JCI "cure rate approached 90%". Periods and measures may differ.
- "Less than 1%" in 1955 comes only from the Cancer History Project.
- Nervous-system relapse and "the edge of life": Wikipedia only; I used it for orientation only.
- Zubrod's malaria analogy: only in the Benz-Nathan piece in JCI.
- Freireich's "total bust" memory: only in the 2001 interview; Zubrod's view is as he tells it.
- MOPP and Hodgkin: ASCO Post 2024 reports 80% complete remission in 43 patients with advanced Hodgkin lymphoma at the NCI and says the protocol was severely criticized within the institution; I did not go into it here.
- Frei: born 21 February 1924, lost his life in May 2013 aged 89 (JCI). Freireich: born 16 March 1927, lost his life on 1 February 2021 aged 93 from COVID-19 (ASCO Post).
- Houston in 1965: Cancer History Project, Wikipedia and ASCO Post agree.
- Turkey series: only the abstract was read; the names of the two centers are not in the abstract.
Counting the causes of cancer beyond one country: the founding of IARC (1965)
Today we followed the birth of a treatment principle. Tomorrow we look at the same year, 1965, but from treatment to cause: the International Agency for Research on Cancer (IARC), set up within the World Health Organization. On Day 125 we saw Doll and Hill's smoking finding, on Day 139 the 1964 Surgeon General report, and yesterday and the day before the traces of a metal and a virus; tomorrow I will ask why the question of "why" stopped being one nation's business. The founding date, founders and first functions will be verified tomorrow.
Author's note: The easy route would have been to say "VAMP was born in 1965 and childhood leukemia was cured". The sources are more scattered: the start swings between 1961 and 1964, the first figures are small and varied, some rest on memories from decades after the events, and cure came only with later drugs, supportive care and long follow-up. VAMP has no single day; it has the slow building of a principle. I wrote in the fact-check box that I did not read the first primary reports.
Sources and Further Reading
- Frei E III, Holland JF, Schneiderman MA, Pinkel D, Selkirk G, Freireich EJ, Silver RT, Gold GL, Regelson W. A comparative study of two regimens of combination chemotherapy in acute leukemia. Blood 1958;13(12):1126-1148. doi:10.1182/blood.v13.12.1126.1126 (abstract read).
- Studies of sequential and combination antimetabolite therapy in acute leukemia: 6-mercaptopurine and methotrexate. Blood 1961;18(4):431-454. doi:10.1182/blood.v18.4.431.431 (abstract read; author list could not be verified).
- Frei E III. Combination cancer therapy: Presidential Address. Cancer Research 1972;32(12):2593-2607 (read; tables garbled in the copy).
- Hardisty RM, McElwain TJ, Darby CW. Vincristine and prednisone for the induction of remissions in acute childhood leukaemia. BMJ 1969;2(5658):662-665. doi:10.1136/bmj.2.5658.662 (abstract read).
- Freireich EJ, Brunet LW. Chapter 10: At the NIH: Vincristine, a cure for childhood leukemia. MD Anderson oral history interview, 30 July 2001. openworks.mdanderson.org (read).
- Benz EJ Jr, Nathan DG. A tribute to Emil Frei III. J Clin Invest 2013;123(8):3188-3189 (read).
- The ASCO Post. Oncology community mourns the death of chemotherapy pioneer Emil J. Freireich. 25 February 2021 (read).
- The ASCO Post. The formative years of medical oncology in the United States: a rough and tumble road, Part 2. 25 September 2024 (read).
- Cancer History Project. Emil Frei III and Emil J Freireich page. cancerhistoryproject.com (read).
- MD Anderson Cancer Center. Making Cancer History: Freireich and VAMP, narrated video transcript, June 2009 (read).
- Güneş AM, Ören H, Baytan B et al. The long-term results of childhood acute lymphoblastic leukemia at two centers from Turkey: 15 years of experience with the ALL-BFM 95 protocol. Ann Hematol 2014;93(10):1677-1684. doi:10.1007/s00277-014-2106-0 (abstract read).
- Mukherjee S. The Emperor of All Maladies: A Biography of Cancer. New York: Scribner, 2010 (general context; the relevant part was not read in this session).
This article is Day 142 of the "365 Days of Cancer History" series and narrates the birth of combination chemotherapy in childhood leukemia as a historical event. It is historical-popular content, not a diagnosis or treatment recommendation. For treatment decisions consult your physician. Sources are verified; details that rest only on a secondary source or a memoir, or that could not be read, are marked in the fact-check notes.
Was this article helpful?
Your feedback helps us improve our content.
