Pharmacological Differences Between Anabolic Steroids

One rather key issue to usage of anabolic-androgenic steroids (AAS) is how one chooses which to use, or which combination to use, and indeed, why combinations might be superior to comparable amounts of single steroids. The issue of combining AAS for most efficient muscle gain is one that has been entirely neglected in the medical literature, since acquisition of muscle is not considered of therapeutic necessity, and observations by bodybuilders have unfortunately generally not been made in a systematic manner. I cannot yet give definitive and complete answers on this matter, but some things are clear at this point, and general support for the principle of synergy can be found in some scientific studies.

Pharmacology in the Simplest Case

First, let us consider the case of the simplest kind of drug. This drug would act in only one way, and would do so by binding to a receptor and activating it. The amount of activity performed by that type of receptor would be directly proportional to the number of receptors that bound the drug. Nothing else of any kind would be going on with this drug.

There might also be similar drugs which worked the exact same way, binding to the same receptor. The only ways in which these drugs could differ from the practical point of view are in pharmacokinetics (how fast each drug enters and clears the body) and how potent each drug is. The latter term is one that may easily be misunderstood due to common usage differing from scientific usage. Potency refers to how little of a drug is required to give a defined amount of effect. For example, if one may obtain the desired therapeutic effect in 50% of subjects with 1 mg/day of Drug A or 100 mg/day of Drug B, then Drug A is 100 times more potent.

This does not mean that Drug A is necessarily better! One can get the same effect from Drug B simply by using 100 times as much. It might be the case that Drug B might be preferable despite the higher required dose: for example, if Drug A leaves the body too quickly or too slowly, or has more toxicity for the same therapeutic effect. It means only that in comparing the drugs, to compare them equally, one must compare the effects of 1 unit of Drug A to 100 unit of Drug B.

To understand this a little more, unfortunately we have to use a little math. One could skip over the math if desired and just look at the conclusions which follow fairly easily from the numbers calculated.

Drugs and receptors interact with each other according to a simple equation:

(conc. of drug) (conc. receptor)
Kd = ————————————-
(conc. of drug«receptor)

where (conc. of drug«receptor) is the concentration or amount per volume of receptors that have drug bound to them, and (conc. of drug) and (conc. of receptor) refer to the concentrations of free drug and receptor respectively.

This number Kd is a constant (always the same) for any given drug, but will vary between drugs of different potencies. This fact allows us to calculate the percentage of receptors occupied if we know Kd and the amount of drug.

Kd will be expressed in units of concentration, for example, 1 nanogram per liter. More potent drugs have lower Kd values. In our comparison of Drugs A and B where A was 100 times more potent, if Drug A had a Kd value with the receptor of 1 ng/L, then drug B would have a Kd of 100 ng/L: you would need 100 times more of Drug B to get the same effect.

What would happen therapeutically if you “stacked” Drug A and Drug B?

You can play with the math and you will find that using blends of A and B, where one keeps in mind that B is 100 times less potent and therefore uses 100 mg of it for each unit of A it replaces, that one gets the exact same result regardless of the stacking. Let’s say that Drug A comes in 1 mg tablets and Drug B comes in 100 mg tablets. Each tablet therefore gives the same effect. In the case of the simplest type of drug such as these two drugs, the effect is identical whether one uses 10 tabs of A, 10 tabs of B, or 5 tabs of each. The same number of receptors are occupied regardless and the effect is the same.

Therefore, stacking these drugs makes about as much sense as stacking two brands of aspirin or two brands of coffee. It is okay if one happens to have both available, but there is no reason to go out and buy the second brand in the hopes that stacking it will give more of a caffeine buzz, or more pain relief.

The mixing might make sense if there were a pharmacokinetic difference: perhaps one of the brands of aspirin is time-released and you want both an instant hit for immediate pain relief as well as sustained action. (The sustained action though could be obtained with only the regular brand, simply by taking small amounts frequently.)

Application to Anabolic Steroids

Now the obvious question here is, Is the same type of drug response true with AAS, or are more complex things going on? Let’s say that, used alone, the same effect is obtained from 1 “Deca-unit” of Deca (let’s say that a Deca-unit is 400 mg) or from 1 “Dianabol-unit” of Dianabol (let’s say that this is 280 mg/week in divided doses every day). If these drugs were as simple in action as Drugs A and B, then the math says that the same result will be obtained regardless of whether one uses one “Deca-unit” of Deca per week, one “Dianabol-unit” of Dianabol per week, or half a unit of each respectively in a stack.

This however is not what happens. Using half a Deca-unit and half a Dianabol-unit per week (say 200 mg/week Deca and 20 mg/day Dianabol) gives better gains than using one unit of either alone. This effect is called synergy and results when there is more than one mechanism of action. The above math remains correct for any given receptor but this is saying that there are more things going on in the body than simply binding to one receptor.

Aside from this and other practical but well-confirmed observations, there is scientific evidence that this is indeed the case.

Scientific Evidence for Multiple Modes of Action

The first thing to consider is whether or not a single mode of action is sufficient to explain all results, as with the simplest case described for Drugs A and B, or whether data is in conflict with such a model.

The equation given earlier allows one, given a measured Kd value, to calculate what percentage of receptors is occupied for a given concentration of drug.

The Kd value for testosterone and the androgen receptor (AR) actually is not known with great precision for humans, but is approximately .44 nmol/L.1 Free testosterone levels in normal men average approximately .07 nmol/L.2,3,4

Contrary to previous statements made by me(although those statements had been made in the scientific literature) this indicates that normal testosterone levels are not sufficient to saturate the AR. The equation given shows that with these values for free testosterone (Tf) and for Kd, one would expect only 14% of ARs to be occupied at any time. Increasing Tf by ten times would improve this to 61% occupancy, which still is not saturated. Increasing twenty times would yield further improvement to 76%. Perhaps this correlates well with the observation that gains improve markedly relative to low dose as one increases amount of testosterone used to 1 gram per week, but going to 2 grams per week offers only a modest further increase.

These results surprise me and are definitely contrary to accepted wisdom. I can only speculate at the moment that those who were trying to determine whether or not receptors are saturated made the mistake of performing the calculation with total testosterone levels instead of Tf. Doing so would lead to that conclusion but is an incorrect method.

I had been going to argue as I had previously that the dose response curve, which extends at least to the 1 gram per week level,5 indicates that there must be more than one mechanism of action, since response increases even past the point of saturation. However these calculations just performed indicate that the dose response curve, through the range that has been studied, is in accord with known values for Kd. This doesn’t prove that there is only one mechanism, but just that one mechanism is not disproven by the dose response curve.

Is there other evidence for multiple mechanisms?


First, there are indisputably molecular targets that are not ARs within some cells which bind androgen and give pharmacological response to androgen. These targets may well have (and in some cases are shown to have) quite different binding properties than the AR does. One AAS might be more potent than another at the AR, but less potent at this other target.

Now these targets are not well known or characterized at all, but there is compelling evidence for their existence. First, as discussed above, for any given target (or receptor) drugs acting only at that receptor will behave the same way and differ only in their potencies. Now if all AAS behaved the same way and differed only in their potencies, and had the same ratios of potency regardless of the activity being studied (whether in muscle or skin or nerves, etc.) then this would be consistent with there being only one target or receptor. However, if some AAS are effective in some activities but do nothing in others, while other AAS do have these other activities, then this can’t all be occurring from the same receptor.

Most of the research in this area is rather far removed from bodybuilding, but the principles still apply. Biochemistry is usually much broader than any one specific cell being studied. (For example, most human biochemistry was actually learned originally by study of E. coli and with later research found to be identical in man.) Thus, while we may not care about ductal branching morphogenesis in the developing rat prostate, the fact that a peculiar biochemical mechanism of androgen response occurs here implies that such a mechanism may well exist in things we are interested in, such as bodybuilding. The possibility at least exists.

Speaking of ductal branching morphogenisis in the developing rat prostate,6 here indeed different steroids behave differently. While to the AR testosterone is less potent than DHT, here the reverse relationship was found. Furthermore, methyltrienolone, which is a more potent agonist (activator) of the AR than is DHT, was no more effective than DHT in inducing ductal branching and was less effective than testosterone. This cannot be explained by assuming that aromatization of testosterone to estradiol contributed to the process, because 5a-androstan-3a,17b-diol (which cannot aromatize) was similarly potent. Thus, there is some target or receptor in these tissues which has different “preferences” (Kd values, and different rank order of potency) than the AR does. Could this also be the case for muscle growth? Perhaps.

Another example is found in the virilization of the mammary gland of female rats.7 The same results are seen here as in the above example of the rat prostate. Testosterone (T) has more activity than DHT does, though at the AR that would not be so.

Differences also are seen in the male accessory glands of the rabbit and rat.8 Testosterone propionate and DHT propionate were found to be equally potent in supporting growth and secretory activity of these glands, but the above-mentioned 5a-androstan-3a,17b-diol was considerably more potent than these in the prostate but completely ineffective in the epidydimis. Furthermore, use of an antiandrogen (AR blocker) did not affect the function of the epidydimis at all. Thus, the activity of testosterone and DHT in this tissue is not via the AR. Are there muscle-building activities that are not via the AR? If the mechanism exists in one tissue it probably does in others as well.

Here is an activity that is itself of more interest: regulation of lipolysis (fat release) in adipocytes (fat cells).9 Testosterone, but not DHT, stimulated catecholamine-induced lipolysis. The findings indicated that T but not DHT induced upregulation of b-adrenergic receptors.

Use of an aromatase inhibitor did not change these results, so conversion to estrogen was not responsible for the difference. If this activity were via the AR, DHT would also have exhibited this effect. Clearly then, something is going on that is not via the AR.

Differential effects of different AAS on human fat cells have also been seen.10 Oxandrolone was most effective in reducing subcutaneous abdominal fat and visceral fat in obese middle-aged men while weight did not change, as a result of muscle mass increase. Testosterone enanthate gave a small decrease in subcutaneous fat but a slight increase in visceral fat. Nandrolone decanoate also increased visceral fat while decreasing subcutaneous fat. If these activities were via the AR, all three steroids should give the same effects, differing only in potency or the dosage required.

There are some interesting studies on sexual receptivity of female rats. Methyltestosterone, methandrostenolone (Dianabol), nandrolone decanoate, and stanozolol all interfered with sexual receptivity (a different result than seen in human bodybuilders) while testosterone propionate did not.11

In male rats,12,13,14 differential activities are also seen. In intact (non-castrated) male rats, testosterone cypionate, nandrolone decanoate, and methandrostenolone (Dianabol) were all able to support male sexual behavior, while methyltestosterone, stanozolol (Winstrol), and oxymetholone eliminated male sexual behavior. Again, these results are different than are seen in human bodybuilders. Testosterone cypionate was able to maintain ejaculation in castrated rats, while oxymetholone (Anadrol) was barely able to do so, and stanozolol was unable to do so. This however might have to do with estrogenic activity – use of an aromatase inhibitor was not tried.

Oxandrolone was found incapable of supporting reproductive development in the young male rat.15 Weight of testes, prostate gland, and seminal vesicles were all below controls, and Leydig cells were severely depleted. Again, it was not ruled out that reduced estrogen levels of the oxandrolone-treated animals might have been to blame, so this does not actually prove a non-AR-dependent mechanism for reproductive development. It does indicate that androgens other than testosterone combined with low estrogen levels can result in fertility problems in the rat, and therefore long-term use of nonaromatizing steroids might affect sperm count in the human as well.

Virilizing activities in female rat fetuses also showed a trend of potencies different from trends of binding affinities to the AR.16 The specific test used was measurement of the abridgment of urovaginal septum length: admittedly not so directly relevant for female bodybuilders. The most active AAS was stanozolol, which was more active than methyltestosterone despite having much poor binding affinity to the AR than that steroid.17

In Syrian hamster embryo cells, trenbolone, a more potent agonist of the AR than testosterone, was found unable to transform these cells while testosterone was effective.26 This indicates that the mechanism cannot be simply via the AR.

The AR is not a membrane-associated receptor, but exists within the cell. However, receptors for testosterone have been found in the cell membranes of T cells. The activity of testosterone (increase of amounts of Ca++within the cell) occurs within seconds (and therefore cannot be via interaction with DNA resulting in increased protein synthesis, since this is a slow process) and was not affected by an AR blocker.18 This effect has also been seen in Sertoli cells.19

Androgen binding receptors have also been found in cell microsomes – these receptors cannot interact with DNA because of their location.20,21,22 Stanozolol has been found to have activity in microsomes that testosterone does not.23,24,25

Lastly, while only stanozolol was tested and therefore we cannot know if there is differential activity between different steroids or not, stanozolol induced a type of skeletal muscle injury that was thought perhaps to stimulate growth, and to induce gene expression by an AR independent mechanism.27 At last, a specific example related to muscle that shows that not all activity is via the AR alone.

We might also speculate that AR upregulation (which has been demonstrated to occur under some conditions (see Androgen Receptor Regulation) is probably not itself mediated by the AR. It would be an unstable mechanism to have the number of ARs increase as a result of increasing numbers of activated ARs. More likely there would be another mechanism.

We may also speculate that different AAS have different effects on nerves, and these effects (being rapid) are not mediated by the AR. E.g., fluoxymesterone, while it binds fairly poorly to the AR, is highly potent in stimulating aggression, and this activity occurs quickly.


What to do with this information? Unfortunately we cannot yet identify how many non-AR-mediated activities there may be. There are I think at least two: activity in microsomes and activities in nerves. There may be more. For example, differentiation of satellite cells of muscle into mature muscle cells might be a non-AR mediated activity.

The practical application of this is that one should not use only a steroid which is good at some things but not others. Examples of this would be Deca and Primobolan (good agonists of the AR but this is not sufficient to make them outstanding anabolics) and Anadrol and Dianabol, which are weaker agonists of the AR yet effective anabolics. Combining drugs of one type with the other is synergistic. It may also be that testosterone and trenbolone are synergistic – trenbolone is much more potent at the AR but (as seen with the Syrian hamster cells) testosterone has at least one activity that trenbolone does not. Winstrol has metabolic properties that testosterone lacks.

Is there a reason to use both Dianabol and Anadrol together? Does one have one non-AR mediated activity which the other lacks? I think not, although Anadrol does seem to have progestogenic activity which Dianabol does not. In any case I don’t know anyone who likes to combine these drugs.

Right now I would say that all bases are covered with testosterone plus trenbolone plus (Dianabol or Anadrol) plus Winstrol. I am not sure that there is no overlap: perhaps the activities of testosterone are covered by the other three.

I hope that future careful observations of results obtained in bodybuilding will allow a more precise answer to this question in the future.

Pharmacological differences between anabolic steroids
Pharmacological differences between anabolic steroids


  1. Hodgins MB.J Steroid Biochem. 1983, 19, 555-9
  2. Diamond T, Lynch W, et al.Cancer. 1998, 83(8), 1561-6
  3. Skarda ST, Burge MR.West J Med.1998, 169, 9-12
  4. Demark-Wahnefried W, Paulson DF.J Androl. 1997, 18(5), 495-500
  5. Forbes GB.Metabolism.1985, 34, 571-3
  6. Foster BA, Cunha GR.Endocrinology.1999, 140, 318-28
  7. Goldman AS, Shapiro B, Neumann F.Endocrinology.1976, 99, 1490-5
  8. Jones R.J Endocrinol.1977, 74, 75-88
  9. Xu X, De Pergola G, Bjorntorp P.Endocrinology.1990, 126, 1229-34
  10. Lovejoy JC, Tulley R, et al.Int J Obes Relat Metab Disord.1995, 19(9), 614-24
  11. Blasberg ME, Clark AS.Horm Behav.1997, 32(3), 201-8
  12. Clark AS, Harrold EV, Fast AS.Horm Behav.1997, 31, 35-46
  13. Clark AS, Barber DM.Physiol Behav.1994, 56, 1107-13
  14. Clark AS, Harrold EV.Behav Neurosci.1997, 111, 1368-74
  15. Grokett BH, Ahmad N, Warren DW.Acta Endocrinol Copenh.1992, 126, 173-8
  16. Kawashima K, Kuwamura T, et al.Endocrinol Jpn.1977, 24, 77-81
  17. Saartok T, Dahlberg E, Gustaffson JA.Endocrinology.1984, 114, 2100-6
  18. Benten WP, Wunderlich F, et al.FASEB J.1999, 13, 123-33
  19. Gorczynska E, Handelsman DJ.Endocrinology.1995, 136, 2052-9
  20. Steinsapir J, Muldoon TG.Steroids.1991, 56(2), 66-71
  21. Steinsapir J.Receptor.1992, 2, 45-76
  22. Steinsapir J, Muldoon TG.J Steroid Biochem Mol Biol.1990, 37, 697-705
  23. Boada LD, Diaz-Chico BN, et al.J Pharmacol Exp Ther.1996, 279, 1123-9
  24. Fernandex L, Boada LD, et al.Pharmacol Toxicol.1995, 77(4), 264-9
  25. Fernandez L, Diaz-Chico BN, et al.Endocrinology.1994, 134, 1401-8
  26. Tsutsui T, Barrett JC, et al.Carcinogenesis.1995, 16, 1329-33
  27. Abu-Shakra SR, Nachtman FC.Ann NY Acad Sci.1995, 761, 395-9

Leave a Reply

Your email address will not be published. Required fields are marked *