Conserving surgery

The resistance to less

For the better part of the twentieth century, the standard for operable breast cancer was William Halsted's radical mastectomy: the entire breast, the pectoralis muscles beneath it, and the axillary lymph nodes cleared in a single block. The operation was built on a theory — that the disease spread outward in an orderly anatomical sequence — and as long as the theory held, taking less seemed like taking a risk. Surgeons who proposed doing exactly that found the burden of proof squarely on them.

Geoffrey Keynes in London was among the earliest to articulate what conservative surgery might look like. Working at St Bartholomew's Hospital in the 1920s and 1930s, he removed tumours with limited margins and added radium needle implants, reporting outcomes he considered comparable to radical removal. His results attracted interest but not consensus; without a controlled comparison, sceptics could always attribute whatever survival he observed to patient selection rather than equivalence.

The impasse was structural. Surgical practice in most specialties changes through accumulated experience and the authority of senior practitioners. Breast cancer in particular had accumulated enormous institutional weight behind the radical approach. Changing it required a kind of evidence that surgery had rarely generated: a properly randomised trial with large numbers, long follow-up, and a primary outcome measured in survival rather than local control.

What the trials showed

Two trials, running in parallel on different continents, eventually supplied that evidence. In Pittsburgh, Bernard Fisher and the NSABP B-06 programme enrolled women with tumours no larger than four centimetres and randomly assigned them to total mastectomy, lumpectomy alone, or lumpectomy followed by breast irradiation. In Milan, Umberto Veronesi led Milan I, which compared the Halsted radical mastectomy directly with a quadrantectomy — removal of the tumour-bearing quadrant of the breast — plus radiotherapy. Both trials opened in the 1970s and reported first results in the early 1980s.

The headline finding in each was the same: survival did not differ between the groups. Fisher's results, published in the New England Journal of Medicine in 1985, showed that lumpectomy with radiotherapy produced equivalent survival to mastectomy, with local recurrence meaningfully higher only in the lumpectomy-alone arm — establishing that radiation after conservative surgery was not optional. Veronesi's group found equivalent overall and disease-free survival with substantially better cosmetic outcomes and less functional disability from the shoulder and arm.

Taken together, the two trials changed what surgery for early breast cancer looked like. They did not argue the case; they measured it. The Early Breast Cancer Trialists' Collaborative Group subsequently pooled data across multiple trials and confirmed the equivalence at scale, embedding breast conservation into the evidence base rather than leaving it as the preference of particular centres.

What conserving means in practice

Conservative surgery is not a single operation. The term covers a range of excisions — lumpectomy, wide local excision, quadrantectomy — distinguished mainly by how much tissue surrounds the tumour. The shared requirement is a clear margin: no tumour cells at the edge of what was removed. Margin assessment is done on the pathology slide; inadequate margins send patients back for re-excision or, ultimately, mastectomy, which is why the pathologist's report sits at the centre of every surgical outcome.

The approach also depends on radiotherapy. The trials that demonstrated equivalence used radiation to the conserved breast as part of the treatment, and subsequent evidence has consistently shown that omitting it increases local recurrence. Whether certain subgroups — older patients with small, low-grade, receptor-positive tumours — might forgo it has been a separate, ongoing question, one that trials have examined without yet closing.

What the trials from Pittsburgh and Milan settled was the larger question: that removing less tissue, properly combined with radiation, does not cost a woman her chance of survival. It took randomisation to establish that, because argument, experience and authority had already been tried, and they had not been enough.