Part X · Metastatic disease: principles and cross-cutting management · Chapter 56

Organ-specific metastases and complications

Which organ-specific decisions rest on a randomised comparison, and which rest on custom.

1 · Bone metastases and skeletal-related events

This chapter covers bone, liver, lung, the serous cavities, the obstructive emergencies and the place of local intervention in disease that is not curable. The central nervous system is the one compartment held apart, because its evidence, its endpoints and its delivery problem are unlike the rest, and it has its own chapter in Central nervous system metastatic disease. The principles that govern all of them are in General principles of metastatic management.

Bone is the commonest first site of distant relapse in hormone receptor-positive disease¹. The colonisation programme behind that is described in Bone, liver, lung, and brain microenvironments compared and in².

The endpoint these trials share deserves inspection before any of their results are quoted. A skeletal-related event is a composite of pathological fracture, radiation to bone, surgery to bone and spinal cord compression. Two of those four are interventions rather than events. The composite therefore measures clinical behaviour as well as disease behaviour. That does not invalidate the endpoint. It does mean the absolute rates travel badly between health systems with different referral thresholds.

The placebo-controlled evidence that established the class is older and sits outside that table. Monthly pamidronate added to chemotherapy reduced skeletal complications in women with lytic bone metastases, and the effect held for at least two years^6,7. Added to hormonal therapy it did the same⁸, and a combined analysis confirmed palliation for at least 24 months⁹. Zoledronic acid then proved more effective than pamidronate in reducing skeletal complications, in patients with at least one lytic lesion^10,11. Intravenous ibandronate improved quality of life¹², and oral and intravenous ibandronate were well tolerated over years^13,14. In ZOOM, moving to 12-weekly zoledronic acid after a year of monthly dosing maintained its effect¹⁵. The denosumab data are reviewed in¹⁶. Osteonecrosis of the jaw is the class risk, and conservative management is preferred to aggressive surgery¹⁷.

Every trial in that table randomised one active bone-modifying schedule against another. That shared design is the finding about the class. Where the two agents met head to head, neither survival nor disease progression separated them. The two interval trials varied the spacing of one drug rather than testing it against nothing. Nothing here says that a bone-modifying agent alters the course of the disease outside the skeleton. These are supportive drugs with a supportive-care justification, however oncological they look on a prescription.

Where the three diverge is in what they measured and in whom. Denosumab versus zoledronic acid measured time to a first skeletal event and tested it for superiority⁵. The two interval trials measured the proportion of patients reaching at least one event inside a fixed window. Each tested that proportion against a non-inferiority margin its investigators chose in advance. A delay in a first event and a proportion with any event inside a window are different objects. The second carries a judgement about what size of difference would matter. Neither design was capable of producing the other's answer.

The agent question rests on the head-to-head, the only randomised comparison of the two drugs against each other. It makes the choice between them a choice between toxicity profiles rather than between disease outcomes. Renal adverse events and acute-phase reactions were more frequent with zoledronic acid. Hypocalcaemia was more frequent with denosumab, which is why calcium is checked after starting it.

The interval question was asked twice, at different moments in a patient's course. CALGB 70604 (Alliance) randomised the interval from the start of treatment and for the whole two years, in three cancers enrolled together³. OPTIMIZE-2 randomised it in breast cancer alone, and only after nine or more monthly doses had already been given⁴. Choosing a twelve weekly interval at the outset and stepping down to it after a year are different decisions. CALGB 70604 addresses the first and OPTIMIZE-2 the second. The habit of starting at the longer interval therefore rests on CALGB 70604. The schedules did not separate within the breast cancer stratum of that trial, which is reassuring rather than powered.

CALGB 70604 found bone turnover less completely suppressed on the longer interval³. That is an assay result. The same trial counted skeletal events. They did not follow the marker. A pharmacodynamic marker that fails to track the clinical endpoint in its own trial is not a reason to shorten the interval.

Three things the table cannot show. None of these trials was built around a comparison with no bone-modifying agent at all, so none of them establishes that the class beats nothing in a patient who already has bone metastases. Every interval result is a zoledronic acid result. No randomised trial has tested a longer denosumab interval, so the twelve weekly habit does not transfer across the class. And the composite gives a radiotherapy referral the same weight as a fracture.

The adjuvant literature makes a different claim and is regularly quoted as though it made the same one. Adjuvant bisphosphonates reduce bone recurrence and breast cancer mortality in postmenopausal women¹⁸. That is prevention of bone metastasis in patients who do not have it. It says nothing about the treatment of established bone disease Treatment toxicity and supportive care. The two bodies of evidence should not be merged.

Bone is also the hardest compartment in which to call a response. Sclerotic change in a healing lytic lesion reads as a new finding on computed tomography. Tracer uptake on scintigraphy reflects osteoblastic activity rather than tumour cell number Response assessment and patterns of progression.

2 · Liver and lung disease

Liver disease is what most often converts a chronic illness into an acute one. The variable that matters is function rather than volume. A patient with extensive parenchymal replacement and normal synthetic function has a different set of options from one with similar imaging and a rising bilirubin. Drug clearance changes with the second and not with the first.

Visceral crisis is defined clinically rather than radiologically, as rapidly progressive visceral involvement with organ dysfunction that requires a rapid response. It is the one situation in hormone receptor-positive disease where chemotherapy has customarily displaced endocrine and targeted therapy.

That custom has been tested. The question those trials were built to answer is argued in Rapid progression and visceral crisis. The trials themselves are tabulated in CDK4/6 inhibition, combinations and sequencing, where the first-line record belongs.

RIGHT Choice is the one that reaches this section. Its result needs stating carefully. It enrolled pre- and perimenopausal women with clinically aggressive disease, and progression-free survival favoured ribociclib with endocrine therapy over combination chemotherapy. Of its 222 patients, 106 were recorded as being in visceral crisis by their own investigator rather than by an adjudicated protocol definition¹⁹. That is a measurement problem before it is a biological one, because investigator judgement is exactly the thing a clinic is trying to calibrate. What the result establishes is that the reflex to chemotherapy is wrong more often than it was assumed to be. It does not establish that it is never right. It does not mark the threshold at which it becomes right.

The practical form of that is a question about time rather than about drug class. Median time to response in RIGHT Choice was 3.2 months with combination chemotherapy and 4.9 months with ribociclib and endocrine therapy. Chemotherapy is the faster of the two. Neither is fast. Where organ function is failing over days, the ordering of those two medians is the whole decision. Where it is declining over months it is not.

Two limits sit on that trial. Neither is a criticism of its conduct. It is a randomised phase 2 with an open-label investigator's choice control. Overall survival has not been reported, so the trade has been measured on progression-free survival alone. And the patient in true visceral crisis with failing organs is excluded from trials of this kind by performance status and organ function criteria. That decision will not acquire a randomised answer, which is the reason for naming it as a judgement rather than dressing it as evidence.

Lung parenchymal metastases are usually asymptomatic until they are extensive, with a tempo slower than liver disease of comparable burden. The decision is therefore rarely forced by the lung itself.

Lymphangitic carcinomatosis is a separate problem with a separate tempo. It is the one to recognise early. Breathlessness out of proportion to the parenchymal findings, with septal thickening on imaging, is attributed first to infection and second to treatment-related pneumonitis. In a patient receiving a deruxtecan-based conjugate that second possibility is urgent, because drug-related interstitial lung disease in this class carries real mortality²⁰. The distinction changes what happens in the next 24 hours and cannot be made on a chest radiograph.

Why particular organs are colonised is developed in Metastatic dissemination and organ tropism, with the lung programme described in²¹ and the vesicle-based account of preferential seeding in²².

3 · Malignant effusions and chest wall disease

A malignant pleural effusion has three management options. The choice between them turns on prognosis and on preference rather than on efficacy in relieving the symptom.

Both trials randomised the same two strategies against each other. They agree on more than they are usually credited with. Breathlessness improved on both arms in both trials. Both found that a catheter reduces the need for a further pleural procedure. Both found more adverse events with the catheter. None of that is in dispute.

They read as disagreeing because they set different primary endpoints. TIME2 set a daily breathlessness score over the weeks immediately after the procedure²⁴. AMPLE set total days in hospital out to death or twelve months²³. A short-horizon symptom score and a year-long count of hospital days cannot contradict each other. The divergence is the design, not the device.

That also decides which trial answers which question. TIME2 is the one that addresses relief of breathlessness. It found no difference between the strategies on its own scale. That is a null result rather than a demonstration of equivalence, because no equivalence margin was set in advance. AMPLE is the one the choice actually rests on. It measured time in hospital over a horizon long enough for the difference to accumulate. Time in hospital is what a patient with months to live is being asked to trade.

TIME2's late breathlessness difference in favour of the catheter is the single result that looks like a device effect. It should be read with care. It was a secondary comparison made months after randomisation, in the patients still alive to complete a score rather than in the population that was randomised.

This is therefore a preference-sensitive decision rather than an efficacy one. It should be offered as such. A patient with months to live and a strong wish for time out of hospital chooses differently from one expecting a longer course.

Three things the table cannot supply. Neither trial enrolled breast cancer alone, so no breast-specific estimate exists for either endpoint. Neither randomised a catheter against repeated therapeutic aspiration, which is what a frail patient often actually receives. And neither reports what managing a drain at home costs the person who does the draining, which is the burden the catheter moves out of the hospital and onto the household.

Pericardial effusion is dangerous for physiological reasons rather than volumetric ones. Tamponade depends on the rate of accumulation and on pericardial compliance. A small rapidly accumulating effusion is more dangerous than a large slow one. Echocardiography answers the question that a computed tomography report describing a moderate effusion does not.

Malignant ascites is managed by paracentesis, and by an indwelling catheter where drainage is repeated, on the same logic as the pleural space. That logic is an analogy rather than a result. Neither space has been randomised. The table above holds nothing outside the pleura.

Chest wall disease produces a symptom burden disproportionate to its extent and disproportionate to the attention it receives. Ulceration, exudate, odour, bleeding and pain occur together and each needs specific management. Radiation and systemic therapy address the tumour. The wound needs dressing selection, odour control and bleeding management. That is a specialist skill. Referring it to tissue viability or wound care services is not an admission that oncology has failed. Treating it as one leaves a manageable problem unmanaged.

4 · Obstructive complications and oncologic emergencies

Metastatic spinal cord compression is the emergency in this disease where hours change the outcome. The variable predicting function afterwards is function beforehand.

The two trials there are not two answers to one question. One asked whether to add decompressive surgery to radiotherapy. The other asked how many fractions of radiotherapy to give. They share a population, metastatic cancer of any primary site. They share an endpoint expressed as walking. They constrain each other in nothing else. Reading them as agreeing or disagreeing is a category error.

Patchell decompressive surgery trial is why a surgical opinion is sought before radiotherapy alone is accepted²⁶. It is load-bearing because it randomised the addition against the thing every such patient was otherwise going to get. Two limits belong with it. Both are about how far the result travels. The trial enrolled metastatic cancer generally rather than breast cancer. It also stopped early at a prespecified interim analysis, which tends to inflate the estimated effect.

SCORAD asked the narrower question. The way it failed is the instructive part. A single fraction missed its non-inferiority margin for walking at week eight, while meeting that margin at weeks one, four and twelve²⁵. A result that holds before a timepoint and again after it is more likely to be a property of the timepoint than of the dose. The explanation is attrition. A large part of the randomised population had died before the week eight assessment, so the primary endpoint was measured in the survivors. Its survival column points the same way. The two schedules did not separate on survival, which is what makes the choice of schedule a logistical decision rather than an oncological one.

Three things neither trial can tell a breast oncologist. Neither enrolled breast cancer alone, so the estimate a reader of this chapter wants does not exist. The patient with hormone receptor-positive or HER2-positive disease and years of expected survival is under-represented in both. Nobody randomised the modern operation, because separation surgery followed by stereotactic radiotherapy has not been compared with either arm of either trial. And nobody randomised corticosteroid dose or the interval from symptom to scan, which are the two variables a physician actually controls in the first hour.

Ambulatory status at presentation carries the same warning. It is the strongest predictor of ambulatory status afterwards. It is a baseline variable rather than a randomised one. No treatment choice can improve it. Only a shorter delay can preserve it.

Superior vena cava obstruction in breast cancer is usually subacute rather than immediately life-threatening. Endovascular stenting relieves it faster than radiation does. The two are commonly combined. The immediate priorities are airway and cerebral oedema rather than the caval anatomy itself.

Hypercalcaemia of malignancy is treated with rehydration and a bone-modifying agent. Its appearance is also information, because it frequently marks a change in disease behaviour. It should prompt reassessment rather than only correction.

Biliary and ureteric obstruction present the same structural question. The decision to stent is a decision about what follows the stent. A stent placed without a subsequent systemic plan buys a laboratory value rather than time.

5 · Local intervention in the metastatic setting

Local treatment in metastatic disease is given for three different reasons. They should never be argued as one.

Symptom control. A painful bone deposit, a bleeding chest wall, an obstructing lesion. The justification is the symptom. No survival claim is required or implied.

Structural protection. An impending fracture, a threatened cord, a compromised airway. The justification is the event being prevented Obstructive complications and oncologic emergencies.

Disease control intent. The oligometastatic and oligoprogressive arguments, where local treatment is given in the hope of changing the course of the disease.

Only the third makes a claim about the course of the disease. That is the claim a randomised comparison against systemic therapy alone can settle.

Both trials asked the same question in form. Does local treatment given for disease control change overall survival, against standard treatment alone? They differ in what was treated, in who was enrolled, and above all in the standard of proof each set itself. That last difference is the one that matters. It is a matter of design rather than a disagreement about whether ablating a deposit helps anybody.

EA2108 is the clearer of the two. It contradicts the stronger intuition. Women presenting with metastatic disease and an intact primary received systemic therapy for four to eight months. Those without progression were then randomised to locoregional treatment of the primary or to continuing systemic therapy alone²⁷. Removing the primary improved local control while improving neither survival nor quality of life. Local control and survival are separate outcomes. Improving the first delivered neither of the others.

That design is what makes it load-bearing. It is the only one of the two conducted in breast cancer. It randomised the comparison a clinic is actually tempted by. And the run-in placed the intuition in its strongest possible position. Randomising only the patients whose disease had not progressed selects for systemic control, which is the setting in which clearing a residual primary should pay if it ever does. It did not pay there either.

SABR-COMET points the other way and is much the weaker design. It was a randomised phase 2 screening study with its significance threshold set at 0.20, against the conventional standard a phase 3 trial holds itself to²⁸. A threshold of that kind is built for deciding what to test next rather than for deciding what to do on Monday. Breast cancer was one of four common primaries in it and a minority of the enrolment, so the breast-specific evidence inside it is thinner than the headline suggests.

There is a measurement problem underneath it as well. Entry required a countable number of lesions, all of them treatable by ablation. That count is a property of the scanner and of the reporting convention as much as of the disease. Randomisation balances indolent biology between the arms. It does nothing to make the result transferable to a patient whose deposits were merely counted differently on a more sensitive study.

Three comparisons nobody has randomised follow from that table. Ablation of metastases has never been tested in breast cancer alone. The patients who progressed during that run-in were never randomised, so the situation in which a primary is left behind in advancing disease is untested. And nobody has randomised local treatment of an oligoprogressive site at all.

Oligoprogression is the one situation where a disease-control rationale rests on a coherent biological argument. A single progressing deposit while everything else responds suggests a resistant subclone in one place, which is the divergence picture in Convergent and divergent evolution and what each implies for sequencing. Treating that site and continuing the systemic regimen follows from the inference. It has still not been shown to extend survival Oligometastatic and oligoprogressive disease.

References

All 30 references

1. Kennecke H, Yerushalmi R, Woods R, et al. Metastatic behavior of breast cancer subtypes. J Clin Oncol 2010 28:3271-3277. PMID 20498394
2. Kang Y, Siegel PM, Shu W, et al. A multigenic program mediating breast cancer metastasis to bone. Cancer Cell 2003 3:537-549. PMID 12842083
3. Himelstein AL, Foster JC, Khatcheressian JL, et al. Effect of longer-interval vs standard dosing of zoledronic acid on skeletal events in patients with bone metastases: a randomized clinical trial. JAMA 2017 317:48-58. PMID 28030702
4. Hortobagyi GN, Van Poznak C, Harker WG, et al. Continued treatment effect of zoledronic acid dosing every 12 vs 4 weeks in women with breast cancer metastatic to bone: the OPTIMIZE-2 randomized clinical trial. JAMA Oncol 2017 3:906-912. PMID 28125763
5. Stopeck AT, Lipton A, Body JJ, et al. Denosumab compared with zoledronic acid for the treatment of bone metastases in patients with advanced breast cancer: a randomized, double-blind study. J Clin Oncol 2010 28:5132-5139. PMID 21060033
6. Hortobagyi GN, Theriault RL, Porter L, et al. Efficacy of pamidronate in reducing skeletal complications in patients with breast cancer and lytic bone metastases. Protocol 19 Aredia Breast Cancer Study Group. N Engl J Med 1996 335:1785-91. PMID 8965890
7. Hortobagyi GN, Theriault RL, Lipton A, et al. Long-term prevention of skeletal complications of metastatic breast cancer with pamidronate. Protocol 19 Aredia Breast Cancer Study Group. J Clin Oncol 1998 16:2038-44. PMID 9626201
8. Theriault RL, Lipton A, Hortobagyi GN, et al. Pamidronate reduces skeletal morbidity in women with advanced breast cancer and lytic bone lesions: a randomized, placebo-controlled trial. Protocol 18 Aredia Breast Cancer Study Group. J Clin Oncol 1999 17:846-54. PMID 10071275
9. Lipton A, Theriault RL, Hortobagyi GN, et al. Pamidronate prevents skeletal complications and is effective palliative treatment in women with breast carcinoma and osteolytic bone metastases: long term follow-up of two randomized, placebo-controlled trials. Cancer 2000 88:1082-90. PMID 10699899
10. Rosen LS, Gordon D, Kaminski M, et al. Long-term efficacy and safety of zoledronic acid compared with pamidronate disodium in the treatment of skeletal complications in patients with advanced multiple myeloma or breast carcinoma: a randomized, double-blind, multicenter, comparative trial. Cancer 2003 98:1735-44. PMID 14534891
11. Rosen LS, Gordon DH, Dugan W, et al. Zoledronic acid is superior to pamidronate for the treatment of bone metastases in breast carcinoma patients with at least one osteolytic lesion. Cancer 2004 100:36-43. PMID 14692022
12. Diel IJ, Body JJ, Lichinitser MR, et al. Improved quality of life after long-term treatment with the bisphosphonate ibandronate in patients with metastatic bone disease due to breast cancer. Eur J Cancer 2004 40:1704-12. PMID 15251160
13. McLachlan SA, Cameron D, Murray R, Tripathy D, Bergström B. Safety of oral ibandronate in the treatment of bone metastases from breast cancer : long-term follow-up experience. Clin Drug Investig 2006 26:43-8. PMID 17163234
14. Pecherstorfer M, Rivkin S, Body JJ, Diel I, Bergström B. Long-term safety of intravenous ibandronic acid for up to 4 years in metastatic breast cancer: an open-label trial. Clin Drug Investig 2006 26:315-22. PMID 17163265
15. Amadori D, Aglietta M, Alessi B, et al. Efficacy and safety of 12-weekly versus 4-weekly zoledronic acid for prolonged treatment of patients with bone metastases from breast cancer (ZOOM): a phase 3, open-label, randomised, non-inferiority trial. Lancet Oncol 2013 14:663-70. PMID 23684411
16. Stopeck A. Denosumab findings in metastatic breast cancer. Clin Adv Hematol Oncol 2010 8:159-60. PMID 20400929
17. Woo SB, Hellstein JW, Kalmar JR. Narrative [corrected] review: bisphosphonates and osteonecrosis of the jaws. Ann Intern Med 2006 144:753-61. PMID 16702591
18. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Adjuvant bisphosphonate treatment in early breast cancer: meta-analyses of individual patient data from randomised trials. Lancet 2015 386:1353-1361. PMID 26211824
19. Lu YS, Mahidin EIBM, Azim H, et al. Final results of RIGHT Choice: ribociclib plus endocrine therapy versus combination chemotherapy in premenopausal women with clinically aggressive hormone receptor-positive/human epidermal growth factor receptor 2-negative advanced breast cancer. J Clin Oncol 2024 42:2812-2821. PMID 38771995
20. Powell CA, Modi S, Iwata H, et al. Pooled analysis of drug-related interstitial lung disease and/or pneumonitis in nine trastuzumab deruxtecan monotherapy studies. ESMO Open 2022 7:100554. PMID 35963179
21. Minn AJ, Gupta GP, Siegel PM, et al. Genes that mediate breast cancer metastasis to lung. Nature 2005 436:518-524. PMID 16049480
22. Hoshino A, Costa-Silva B, Shen TL, et al. Tumour exosome integrins determine organotropic metastasis. Nature 2015 527:329-335. PMID 26524530
23. Thomas R, Fysh ETH, Smith NA, et al. Effect of an indwelling pleural catheter vs talc pleurodesis on hospitalization days in patients with malignant pleural effusion: the AMPLE randomized clinical trial. JAMA 2017 318:1903-1912. PMID 29164255
24. Davies HE, Mishra EK, Kahan BC, et al. Effect of an indwelling pleural catheter vs chest tube and talc pleurodesis for relieving dyspnea in patients with malignant pleural effusion: the TIME2 randomized controlled trial. JAMA 2012 307:2383-2389. PMID 22610520
25. Hoskin PJ, Hopkins K, Misra V, et al. Effect of single-fraction vs multifraction radiotherapy on ambulatory status among patients with spinal canal compression from metastatic cancer: the SCORAD randomized clinical trial. JAMA 2019 322:2084-2094. PMID 31794625
26. Patchell RA, Tibbs PA, Regine WF, et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. Lancet 2005 366:643-648. PMID 16112300
27. Khan SA, Zhao F, Goldstein LJ, et al. Early local therapy for the primary site in de novo stage IV breast cancer: results of a randomized clinical trial (EA2108). J Clin Oncol 2022 40:978-987. PMID 34995128
28. Palma DA, Olson R, Harrow S, et al. Stereotactic ablative radiotherapy for the comprehensive treatment of oligometastatic cancers: long-term results of the SABR-COMET phase II randomized trial. J Clin Oncol 2020 38:2830-2838. PMID 32484754
29. Priedigkeit N, Hartmaier RJ, Chen Y, et al. Intrinsic subtype switching and acquired ERBB2/HER2 amplifications and mutations in breast cancer brain metastases. JAMA Oncol 2017 3:666-671. PMID 27926948
30. Fitzpatrick A, Iravani M, Mills A, et al. Assessing CSF ctDNA to improve diagnostic accuracy and therapeutic monitoring in breast cancer leptomeningeal metastasis. Clin Cancer Res 2022 28:1180-1191. PMID 34921020